interpreting prehistoric patterns: site-catchment …/67531/metadc4678/m2/... · granted permission...
TRANSCRIPT
APPROVED: C. Reid Ferring, Major Professor and Chair of
the Department of Geography Lisa Nagaoka, Minor Professor Miguel Acevedo, Committee Member Sandra L. Terrell, Dean of the Robert B.
Toulouse School of Graduate Studies
INTERPRETING PREHISTORIC PATTERNS: SITE-CATCHMENT ANALYSIS IN THE
UPPER TRINITY RIVER BASIN OF NORTH CENTRAL TEXAS
Marikka Lin Williams, B.F.A.
Thesis Prepared for the Degree of
MASTER OF SCIENCE
UNIVERSITY OF NORTH TEXAS
December 2004
Williams, Marikka Lin, Interpreting Prehistoric Patterns: Site Catchment Analysis
in the Upper Trinity River Basin of North Central Texas. Master of Science (Applied
Geography), December 2004, 301 pp., 25 tables, 92 illustrations, references, 93 titles.
Archaeologically site-catchment analysis produces valuable information regarding
prehistoric subsistence strategies and social organization. Incorporating archaeological
data into catchment analyses is an effective strategy to develop regional models of
prehistoric site selection and settlement patterns. Digital access to data permits the
incorporation of multiple layers of information into the process of synthesizing regional
archaeology and interpreting corresponding spatial patterning. GIS software provides a
means to integrate digital environmental and archaeological data into an effective tool.
Resultant environmental archaeology maps facilitate interpretive analysis. To fulfill the
objectives of this thesis, GIS software is employed to construct site-catchment areas for
archaeological sites and to implement multivariate statistical analyses of physical and
biological attributes of catchments in correlation with assemblage data from sites.
Guided by ecological, anthropological and geographical theories hypotheses testing
evaluates patterns of prehistoric socio-economic behavior. Analytical results are
summarized in a model of prehistoric settlement patterns in North Central Texas.
ii
Copyright 2004
by
Marikka Lin Williams
ACKNOWLEDGEMENTS I am very thankful to many people who were instrumental in the process of
completing my thesis. First and foremost, I would like to thank God, my parents, my
siblings, my friends and a number of musicians who have provided continuous
emotional, as well as, spiritual support, throughout the years. Thank you with my
eternal love and respect.
I am also very thankful to all of my teachers who have provided academic support
throughout my education. While I cannot thank all who touched my life and enriched
my education there are several that I would like to mention who have been particularly
influential. Thanks to the first schoolteacher who made an impact on my life during
my early childhood, Mrs. Reckon. Thanks to Dr. Manning from High School who helped
me to open my mind and expand my horizons. Thanks to Dr. Kim and Dr. Beason from
Stephen F. Austin State University. Thanks to Dr. Lindsey and Marilyn Patton, from the
Willis Library, who have been supportive, throughout my education at the University of
North Texas. Thanks to the people from the UNT Registrars office and the Financial Aid
Office, who kept me on board after I graduated from UNT with my Bachelor’s Degree.
Thanks to Ellen Turner, Dr. Thomas Hester and Rowman and Littlefield Publishers who
granted permission to use projectile point illustrations from A Field Guide to Stone
Artifacts of Texas Indians. Finally, thanks to the UNT Geography Department including
Dr. Lyons, Tami Deaton, Dr. Schoolmaster, Bruce Hunter, Dr. Williams, Dr. Acevedo, Dr.
Nagaoka and Dr. Ferring. Thanks to all of the people who have contributed their
knowledge toward completion of this research. You are immensely appreciated.
iii
TABLE OF CONTENTS ACKNOWLEDGMENTS iii LIST OF TABLES viii LIST OF FIGURES x CHAPTER 1: INTRODUCTION 1 Research Orientation 2 Objectives 15 CHAPTER 2: METHODOLOGICAL FOUNDATIONS 18 Cultural Ecology 18 Geographical Theory 22 Site Catchment Analysis 24 Site Catchment Analysis Case Studies 29 CHAPTER 3: RESEARCH SETTING 36 Geology of North-Central Texas 36 Topography 39 Drainage Systems 39 Physiographic Setting 42 North-Central Texas Soils 46 Floral Resources in North-Central Texas 53 Faunal Resources in North-Central Texas 61 Climate of North-Central Texas 71 Past Environments 75
iv
CHAPTER 4: CULTURAL CHRONOLOGY OF NORTH CENTRAL TEXAS 77 Paleoindian (11,500-8,500 BP) 83 Paleoindian Settlement Patterns (11,500-8,500BP) 88 Archaic (8,500-1,250 BP) 89 Early Archaic (8,500-6,000 BP) 93 Middle Archaic (6,000-3,500 BP) 96 Late Archaic (3,500-1,250 BP) 99 Archaic Climatic Change (8,500-1,250 BP) 103 Archaic Subsistence Patterns (8,500-1,250 BP) 104 Early Archaic Settlement Patterns (8500-6000BP) 105 Middle Archaic Settlement Patterns (6000-3500BP) 106 Late Archaic Settlement Patterns (3500-1250BP) 107 Late Prehistoric (1250-250BP) 108 Late Prehistoric Climatic Conditions 113 Late Prehistoric Subsistence Patterns 113 Late Prehistoric Settlement Patterns 115 Unanswered Questions 116 CHAPTER 5: RESEARCH METHODOLOGY 118 Soils as a Diagnostic Tool 124 Soils Variables 125 Soil Settings 126 Soil Potential for Plants and Wildlife 129
v
Interpretations: Potential Economic Value of Soils 130 Site Catchment Soils Diversity 136 Geological Variables 137 Physiographic Variables 139 Climatic Variables 142 Archaeological Variables 144 Site Catchment Analysis of Prehistoric Sites in North-Central Texas 154 Hypotheses Testing 158 Study Area 165 CHAPTER 6: RESULTS 168 Geological Setting 168 Physiographic Setting 173 Topographic Setting 178 Soils: Great Groups 183 Soils: Soil Settings 189 Soils: Economic Interpretations 191 Site Catchment Diversity 204 Climatic Interpretations 206 Site Activities and Density 208 Statistical Conclusions 216 CHAPTER 7: NORTH CENTRAL TEXAS SETTLEMENT PATTERNS 221 Prehistoric Populations along the West Fork of the Trinity 221
vi
Prehistoric Populations along the Elm Fork of the Trinity 227 Prehistoric Populations along the East Fork of the Trinity 235 Functional Site Types in North-Central Texas 238 Prehistoric Adaptations 240 CHAPTER 8: DISCUSSION AND FUTURE RECOMMENDATIONS 246 Discussion 246 Future Recommendations 248 APPENDIX 253 GIS GLOSSARY 293 REFERENCES 294
vii
LIST OF TABLES Table 1 Various chronologies developed for cultural subdivisions 80 Table 2 Soil taxonomy classifications in study area 128 Table 3 Weighting procedure for plant and wildlife estimates 129 Table 4 Geological formations associated with site catchment areas 138 Table 5 Site catchment hypotheses 162 Table 6 Site function hypotheses 163 Table 7 Economic values hypotheses 164 Table 8 Statistical results summary 217 Table 9 Statistical results summary 218 Table 10 Statistical results summary 219 Table A1 Raw data for hypothesis testing 278 Table A2 Raw data for hypothesis testing 279 Table A3 Raw data for hypothesis testing 280 Table A4 Raw data for hypothesis testing 281 Table A5 Raw data for hypothesis testing 282 Table A6 Raw data for hypothesis testing 283 Table A7 Raw data for hypothesis testing 284 Table A8 Raw data for hypothesis testing 285 Table A9 Economic values: Vegetation scores for soil settings 286 Table A10 Economic values: Vegetation scores for soil settings 287
viii
Table A11 Economic values: Vegetation scores for soil settings 288 Table A12 Economic values: Vegetation scores for soil settings 289 Table A13 Economic potential of site catchment areas 290 Table A14 Vegetation potential of site catchment areas 291 Table A15 Wildlife potential of site catchment areas 292
ix
LIST OF FIGURES Figure 1 Lakes and major cities in north-central Texas 3 Figure 2 Geology of north-central Texas 38 Figure 3 Hydrological networks of north-central Texas 40 Figure 4 Physiographic regions of Texas and north-central Texas 43 Figure 5 Soil distribution in north-central Texas study area 47 Figure 6 Archaeological faunal distribution from excavated sites in study area 67 Figure 6a Fauna recovered from archaeological sites in study area 67 Figure 6b Archaeological fauna and change through time 68 Figure 6c Archaeological fauna and change through time 69 Figure 7 Precipitation patterns in Texas and north-central Texas 72 Figure 8 Average monthly temperature and precipitation trends in north Texas 73 Figure 9 Average monthly temperature and precipitation trends comparisons 74 Figure 10 Distribution of prehistoric sites in Texas and north-central Texas 78 Figure 11 Generalized projectile point chronology in north-central Texas 81 Figure 12 Chronological overview of projectile point types 82 Figure 13 Distribution of prehistoric sites in Texas and north Texas 84 Figure 14 Paleoindian projectile point chronology 86 Figure 15 Distribution of Archaic sites 90 Figure 16 Archaic projectile points in north-central Texas 92 Figure 17 Distribution of Early Archaic sites 94 Figure 18 Early Archaic projectile points 95
x
Figure 19 Distribution of Middle Archaic sites 97 Figure 20 Middle Archaic projectile points 98 Figure 21 Distribution of Late Archaic sites 101 Figure 22 Late Archaic projectile points 102 Figure 23 Distribution of Late Prehistoric sites 109 Figure 24 Late Prehistoric projectile points 111 Figure 25 Conceptual diagram of data included in site catchment analysis 122 Figure 26 Procedure to link economic value to site catchment 135 Figure 27 Relative abundance of plant family in study area from Ferndale Bog 143 Figure 28 Diagram of GIS data processing design in site catchment analysis 155 Figure 29 Conceptual diagram of GIS in relation to components of the research 156 Figure 30 North central Texas research study area 166 Figure 31 North central Texas research study area geology 170 Figure 32 Geological settings and time periods 171 Figure 33 Geological settings and occupation frequency 172 Figure 34 Physiographic settings in north-central Texas study area 175 Figure 35 Physiographic settings and time periods 176 Figure 36 Physiographic settings and occupation frequency 177 Figure 37 Topographic settings and time periods 179 Figure 38 Topographic settings and occupation frequency 180 Figure 39 Elm Fork site catchment elevations 181 Figure 40 West Fork site catchment elevations 182
xi
Figure 41 Elm and West Fork great group distribution and site catchments 184 Figure 42 East Fork great group distribution and site catchments 185 Figure 43 Great groups and time periods 186 Figure 44 Great groups and occupation frequency 188 Figure 45 Soil settings and time periods 189 Figure 46 Soil settings and occupations frequency 190 Figure 47 Economic values and prehistoric components 192 Figure 48 Economic values and occupation frequency 194 Figure 49 Economic values and time periods 195 Figure 50 Vegetation potential and prehistoric components 197 Figure 51 Vegetation potential and occupation frequency 198 Figure 52 Vegetation potential and time periods 199 Figure 53 Wildlife potential and prehistoric components 201 Figure 54 Wildlife potential and occupation frequency 202 Figure 55 Wildlife potential and time periods 203 Figure 56 ANOVA of site catchment diversity and occupation frequency 204 Figure 57 ANOVA of site catchment diversity and change through time 205 Figure 58a Distribution of climatically abundant vegetation families 206 Figure 58b Distribution of climatically abundant vegetation families 207 Figure 59 Prehistoric activities and time periods 209 Figure 60 Prehistoric activities and occupation frequency 210 Figure 61 Prehistoric activities, site density and geological settings 211
xii
Figure 62 Prehistoric activities, site density and soil settings 212 Figure 63 Prehistoric activities, site density and great groups 213 Figure 64 Prehistoric activities, site density and topographic settings 214 Figure 65 Prehistoric activities, site density and physiographic settings 215 Figure 66 Prehistoric populations along the West Fork of the Trinity 226 Figure 67 Prehistoric populations along the Elm Fork of the Trinity 234 Figure 68 Prehistoric populations along the East Fork of the Trinity 237 Figure A1 Wildlife potentials and Prehistoric sites 254 Figure A2 Wildlife potentials and Archaic projectile points 255 Figure A3 Wildlife potentials and Early Archaic projectile points 256 Figure A4 Wildlife potentials and Middle Archaic projectile points 257 Figure A5 Wildlife potentials and Late Archaic projectile points 258 Figure A6 Wildlife potentials and Late Prehistoric projectile points 259 Figure A7 Wildlife potentials and Late Prehistoric I projectile points 260 Figure A8 Wildlife potentials and Late Prehistoric II projectile points 261 Figure A9 Vegetation potentials and Prehistoric sites 262 Figure A10 Vegetation potentials and Archaic projectile points 263 Figure A11 Vegetation potentials and Early Archaic projectile points 264 Figure A12 Vegetation potentials and Middle Archaic projectile points 265 Figure A13 Vegetation potentials and Late Archaic projectile points 266 Figure A14 Vegetation potentials and Late Prehistoric projectile points 267 Figure A15 Vegetation potentials and Late Prehistoric I projectile points 268
xiii
Figure A16 Vegetation potentials and Late Prehistoric II projectile points 269 Figure A17 Economic potentials and Prehistoric Sites 270 Figure A18 Economic potentials and Archaic projectile points 271 Figure A19 Economic potentials and Early Archaic projectile points 272 Figure A20 Economic potentials and Middle Archaic projectile points 273 Figure A21 Economic potentials and Late Archaic projectile points 274 Figure A22 Economic potentials and Late Prehistoric projectile points 275 Figure A23 Economic potentials and Late Prehistoric I projectile points 276 Figure A24 Economic potentials and Late Prehistoric II projectile points 277
xiv
CHAPTER 1
INTRODUCTION
Archaeological sites are part of a human ecosystem built within a dynamic living
physical landscape. It is within this interactive matrix that prehistoric communities
interacted spatially, socially and economically. The nature of this matrix is typically
extrapolated from subsistence activities and settlement patterns. Evidence of
subsistence activities is primarily recovered from archaeological sites in the form of
fauna (i.e. animal remains), features (e.g. fire-cracked rock, hearths, middens) and
artifacts (e.g. projectile points, other stone tools, grinding stones, ceramics). In order
to facilitate pattern recognition and assist in the modeling of prehistoric settlement
patterns, archaeologists have examined the characteristics of the micro-environments
that surround site locations. However, broad descriptive generalizations based on
macro-environmental correlations are often found in the literature,
The study of prehistoric sites is not usually complemented by some treatment of their setting, the situation in their immediate vicinity—the principal concern of the inhabitants—tends to be neglected or at any rate overshadowed by generalized statements regarding the physiographic, vegetational, climatic or kindred zones of which they form a part (Vita-Finzi and Higgs, 1970:1).
This is particularly the case for the state archaeological site files housed at the Texas
Archaeological Resource Laboratory (TARL) in Austin. One of the most significant
components missing from state archaeological site files is the specific environmental
context within which these sites and artifacts were discovered. Records housed in state
and university laboratories are increasing in number, yet often site report content is not
substantial enough to facilitate comprehensive regional studies. Glimpses into
1
prehistory that have been provided by archaeological discoveries must be viewed
cumulatively in order to understand the larger patterns that emerge regarding
prehistoric behavior. Although many contract, academic and published reports from the
region make reference to soils, vegetation, geology and zonal physiographic
characteristics of the landscape, there is a general disconnection between the sites, the
data and the interpretive inter-relationships. This research seeks to bridge the gap by
providing a series of maps with interpretive visual representations that provide a bird’s
eye view and facilitate pattern recognition toward developing a more comprehensive
view of site distribution within the environment. Tabular data coupled with interpretive
discussion based on spatial and statistical analysis is used to explain these patterns.
Research Orientation
A substantial amount of research has been conducted, in north-central Texas, with
respect to prehistoric socio-economic systems (Dawson and Sullivan, 1969; Hays et al.,
1972; Bousman and Verrett, 1973; Nunley, 1973; McCormick et al., 1975, Filson et al.,
1975; McCormick, 1976; Lynott, 1977, 1981; Skinner and Baird, 1985; Lebo and Brown,
1990; Prikryl, 1990; Ferring and Yates, 1986, 1997, 1998). These archaeological reports
involving cultural-ecology, subsistence and settlement pattern studies have provided a
wealth of empirical data. Much of the contract and academic research in this region,
conducted between 1968 and 1998, has resulted in the development of models that
discuss site function, distribution and regional settlement patterns. It is important to
note that the research conducted in north-central Texas (figure 1) has revolved around
contract work associated with the construction of reservoirs. Along the Elm Fork of the
2
Trinity River extensive professional archaeological investigations have been conducted.
Research along the East Fork is based on academic and professional efforts but to a
much lesser extent than the Elm Fork. As a result, there is a strong bias in site
distribution related to the intensity of investigations along the Elm Fork. West Fork
sites were primarily reported by amateur archaeologists. The lack of professional
investigations along the West Fork partially explains the paucity of sites. However, the
sites that do exist offer valuable comparative environmental data for site catchment
analysis (SCA).
Figure 1. Lakes and major cities in north-central Texas.
3
Although a variety of regional models have been proposed for this “region”, many of
these models are built in relation to specific excavation sequences typically located in
isolated microenvironments, which are considered to be representative of the region.
General physiographic characteristics are frequently employed to fill in the gaps in order
to extend the model beyond the immediate study area.
Previous studies conducted along the Elm Fork of the Trinity River have resulted in a
variety of interpretations of prehistoric settlement patterns. This section will review the
settlement pattern theories that have been formulated based on prior research in the
region. Several investigators have referred to this region as "marginal" (Lynott, 1977;
Skinner and Baird, 1985; Prikryl, 1990; Story, 1990). However, other investigators such
as Dawson and Sullivan (1969: 5) consider the grassland, woodland, riverine and
transitional ecotonal settings to be “highly opportune” for prehistoric occupation.
Dawson and Sullivan acknowledge the economic potential of the variety of resources
within each environment. Within the grassland, a variety of vegetation and faunal
resources are accessible. Riverine environments, nested within the woodlands, provide
a substantial supply of protein associated with fishing and gathering activities.
Throughout the entire region, small fauna, supplementary foods, medicine, lithics, wood
and various sources for construction can be obtained (Dawson and Sullivan, 1969).
Since multiple environments are present in north-central Texas these authors speculate
that prehistoric groups utilized a generalized and vast array of resources. A highly
mobile and semi-sedentary model of occupation is posited within all three settings.
Based on the character of the environment they propose that food could be obtained
4
during extremely dry or wet climates if groups remained small and spread out across
the landscape. A scattered settlement pattern would sustain balance with climatic and
seasonal fluctuations. Differences in site positioning between Archaic ridge sites and
bottomland Late Prehistoric sites, on the East Fork of the Trinity, are considered to be a
intentional selection of site location geared towards maintaining this balance (Dawson
and Sullivan, 1969).
A food surplus, stored during the spring and summer, would probably be required for group survival during the late winter and early spring. At that time, the probable high water, hibernation of many woodland animals and the scattering of grassland herds would vastly reduce food production. The usual size of the surplus would probably determine functional group size. This surplus could be achieved by hunting and gathering groups from herd slaughter, grass seeds and gathered woodland and lowland products (Dawson and Sullivan, 1969: 6).
Fundamental assumptions of their settlement pattern theory are that “site positioning
reflects basic needs that vary according to cultural adjustment; and that sites are
located near the major source of food” (Dawson and Sullivan, 1969: 2). Given that
several food sources are available, location is thought to have been based on situating
women and children near areas where gathered food sources are maximized (Dawson
and Sullivan, 1969). Essentially, Dawson and Sullivan conclude that prehistoric
population size was primarily dependent on the environment in correlation with the
seasons and the climate. As a result, they propose that an environmentally based
economic threshold was established that dictated prehistoric group size, site selection
and organization.
Hays et al. (1972) identify three distinctive geomorphic locations related to
archaeological sites in an environmental impact study conducted along the Elm Fork of
5
the Trinity. Floodplains, terraces, and uplands are considered to be preferred locations
for archaeological sites (Hays et al., 1972). Floodplain sites investigated by Hays and
others primarily consist of concentrations of burned rock, mussel shells, flint tools and
debitage. According to these investigators several floodplain sites seem to represent
single events of preparing and eating mussels. “Projectile points are present and
indicate this activity may have resulted from a small hunting party stopping for a meal
or for overnight” (Hays et al., 1972: 18). Terrace sites are reported to predominantly
contain lithic scatter and upland sites are typically found where natural quartzite and
chert deposits occur on the surface (Hays et al., 1972). Prehistoric populations
probably quarried raw material at upland locations given the association with local raw
materials (Hays et al., 1972). Faunal remains of both large and small species indicate
hunting in the floodplain and uplands. Remains from the sites, investigated during their
survey, indicate utilization of a variety of microenvironments (Hays et al., 1972).
Overall, Hayes categorizes sites as representing activities such as food preparation
and/or camping, short term seasonal occupations or long term base camps sites that
provided access to various environmental resources.
During an archaeological reconnaissance of the proposed Aubrey reservoir
Bousmann and Verrett (1973) define six micro-environmental zones on the basis of
correlated biology, geology and physiography. These zones include: “rivers and
drainages; floodplains; floodplain rises; fluvatile terraces; upland slopes; and uplands”
(Bousmann and Verrett, 1973: 7). Their purpose for developing these categories is to
provide an explanatory mechanism for inter- and intra-site variation (Bousmann and
6
Verrett, 1973). Employing this model, Bousmann and Verrett (1973: 12) hypothesize
that “the Eastern Cross Timbers district was utilized more often for base camps, and the
prairies were utilized for specialized activities, i.e. hunting.” Supporting evidence for
this hypothesis includes the “presence of permanent water sources, concentration of
wood, highly permeable sandy soils and a stable supply of native plants and animals in
the Cross Timbers” (Bousmann and Verrett, 1973: 12). Another hypothesis, formulated
by these investigators, is that fluvatile terraces specifically function most often as base
camp sites while other zones are occupied only when seasons or climatic conditions
permitted and provided sufficient returns (Bousmann and Verrett, 1973). The central
location of terraces maximized access to all of the resource zones and proximity to
permanent water sources in areas that would be protected from flooding (Bousmann
and Verrett, 1973). In relation to this supporting environmental evidence it is posited
that prehistoric groups moved to upland camps during wetter seasons and lowland sites
during drier seasons. Deviations from this model are expected due to variations in
social organization, religious factors, taboos and other socially defined parameters.
McCormick et al. (1975) proposes that this region served as an avenue of seasonal
migration due to the increased availability of eco-tonal resources. Based on a survey
project conducted along the southern reaches of the Elm Fork these investigators
observe that Archaic seasonal camps are typically located in eco-tonal settings and
secondary activity-specific sites are situated in relation to specific outcroppings, plants,
and water (McCormick et al., 1975: 41). McCormick (1975) posits that Late Prehistoric
groups followed similar patterns, producing seasonal camps and activity specific sites,
7
until agriculture was introduced and large base camps or permanent villages emerged.
According to McCormick’s settlement model, groups seasonally migrate with the bison
and return north in spring to spend the summer months in large agricultural villages.
During the late fall, family sized groups would leave the base camps to follow the bison herds south into the central Texas area, by following the Cross Timbers from the Red River south to the vicinity of Waco, Texas. Temporary camps, showing signs of both male and female activities, should be located well into the Cross Timbers and out on the prairies in association with whatever resource was being exploited. These will be fewer in number than during the Archaic period because of the transitory nature of the usage (McCormick et al., 1975: 41).
The proposed settlement model divides sites into Archaic and Neo-American (Late
Prehistoric) subdivisions. These subdivisions are classified into three groups:
permanent village, temporary campsites and activity-specific sites. In the study area,
located south of Lewisville, along the Elm Fork of the Trinity, this model is found to be
useful. As predicted the large village type settlement is also present in the study area.
Temporary campsites for both the Archaic and Neo-American periods are situated in
similar ecological microenvironments. Permanent villages and activity-specific sites,
such as manufacturing stations, are not recognized within their study area. However,
these site types are found to the southwest further into the Cross Timbers. According
to the model, temporary camps of Neo-American and Archaic groups should be located
in similar areas with affinities in site structure with the exception of diagnostic tools
such as large “dart” points associated with the Archaic and small "arrow" points with
ceramics which identify the Neo-American groups (McCormick et al., 1975: 45). All of
the Neo-American sites are ephemeral and situated along the upland edge, providing
immediate access to a maximum of resource zones. Seasonal Archaic sites tend to
8
display more “permanent characteristics” since these sites were apparently occupied for
a longer duration (McCormick et al., 1975: 45). McCormick and others hypothesize that
the zone of movement was along the prairie-cross timbers boundaries attributed to the
presence of more “permanent” sites (Prikryl, 1990: 31). McCormick et al. (1975) find
Lewisville sites located respectively north and south in almost identical environmental
situations. What is considered to be unusual is the “apparent similarity of activity
regardless of temporal considerations” (McCormick et al., 1975: 78).
The amount of patterning in artifacts and activity areas within a site will increase in proportion to an increase in occupation time. Neo-American permanent villages will display the most evidence of "city planning", Archaic seasonal camps next, then Neo-American transitory camps, and lastly, both Neo-American and Archaic activity-specific stations. In the absence of diagnostic artifacts it is difficult to distinguish Archaic from Neo-American stations where similar activities took place (McCormick et al., 1975: 46).
Reoccupation of sites over a long period of time is also considered to be common in this
region. The settlement pattern exhibited here is thought to have been established
during the Paleoindian and continued until the “westward expansion of the Europeans
altered the native cultures to such an extent that they could no longer continue their
normal settlement and subsistence patterns” (McCormick et al., 1975: 78).
Lynott (1977) presents a regional model for future archaeological research in north-
central Texas and provides an extensive listing of previous investigations in the prairie-
cross timbers area of the region. Lynott (1977) emphasizes subsistence-settlement
models and encourages use of general systems theory. His regionally oriented design
hypothesizes changes in adaptive strategies through time. During the Early to Middle
Archaic, Lynott (1977) observes a shift from open-land to big game hunting to
9
bottomland riverine hunting and gathering. An “increase of population density and
decrease in territoriality” defines the Late Archaic (Lynott, 1977: 158). As a result,
Lynott (1977: 158) proposes a “restricted wandering” settlement pattern where sites
are situated in terraces settings near water and bottomland resources. During the first
phase of the Late Prehistoric, Lynott proposes that the initial stages of tribalization,
sedentism and horticulture were related to an increase in population pressure that
instigated territorial conflicts (Lynott, 1977: 159). The concept of tribalization is related
to the presence of ceremonial structures that emerged on the East Fork of the Trinity.
This situation is thought to have continued to a greater extent during the second phase
of the Late Prehistoric as the “population grew and social boundaries became more
defined” (Lynott, 1977: 160). Throughout time Lynott emphasizes the utilization of
bottomland resources. Regionally, he proposes use of the Wichita model of settlement
and subsistence patterns. This model is based upon a nomadic winter bison hunt in the
prairies with semi-permanent villages occupied during the spring and summer (Lynott,
1977). An alternative model is proposed by associations with the Tonkawa that
suggests seasonal hunting and gathering in north-central Texas (Lynott, 1977). In
general, the settlement pattern that Lynott (1977) proposed for sites revolves around
large bottomland base camps and smaller, seasonally occupied, special activity sites.
Skinner and Baird (1985: 5-11) propose a “marginal territory” model for the Ray
Roberts Lake area, along the northern branch of the Elm Fork. This model is based on
the paucity of sites, small site size, low density of cultural remains at individual sites
and numerous natural environmental variables. It operates on the premise that the
10
diverse resources in the region served as specialized ecosystems. In particular, the
mast resources provided in the uplands of the Cross Timbers and the riparian species,
documented by archaeological reports, reflect these specialized ecosystems. Field
observations of maintenance and extraction sites reveal a concentration in the Cross
Timbers and riparian environmental zones (Skinner and Baird, 1985). Conclusions
based on research conducted at these sites are made according to a demographic
version of a settlement model known as the central based wandering (CBW) community
pattern. The “CBW model”, developed by Beardsley and others in 1956, has been used
as an “organizational tool in various parts of Texas and surrounding regions” (Skinner
and Baird, 1985: 5-1). It is employed as a means of “categorizing sites which had a
limited number of surface artifacts” and where diagnostic tools or pottery are not found
(Skinner and Baird, 1985: 5-1). Due to the lack of datable artifacts, sites are
categorized into functional types on the basis of the character of lithic debris present on
the surface of sites and artifact assemblages. It was expected that by correlating these
two classes of information, consistent site types could be defined and could be
replicated by others (Skinner and Baird, 1985: 5-3).
Based on the results of the site survey, it is noted that 40 of the prehistoric sites
appear to represent a single occupation period while 22 have artifacts from more than
one occupation period. Twenty-one morphological site types are defined by cluster
analysis of the survey data and then consolidated into 7 functional sites (Skinner and
Baird, 1985). Results indicate that of the 117 sites surveyed 62 were datable, 23 were
base camps, 32 seasonal camps, 16 hunting camps, 10 musselling camps, 9 hunting
11
stations, 2 collecting stations and 21 lithic procurement sites (Skinner, 1985).
According to Skinner (1985), undated sites typically include locations where stone was
gathered, or quarried, and where initial tool making processes were carried out.
Skinner and Baird (1985) pay particular attention to site type, age, location, and
other variables such as assemblage, deposition and preservation. Skinner and Baird
(1985) also provide an alternative model of settlement that utilizes the work of others in
the surrounding areas for assistance in the reconstruction of the effective natural
environment. This alternative settlement model associates cultural change and shifts in
settlement locations with changes in the natural environment. It is noted that emerging
patterns indicate that certain locations were preferred over time. According to Skinner
and Baird (1985), while repeated reoccupation may distort functional differences, this
situation provides a means of observing the use of a particular site location during its
lifetime. However, besides reference to the mast crops, there is no specific discussion
as to why these locations were preferred and if there are patterns in particular
environmental variables that made them desirable locations for occupation. There is
also a lack of clarity as to the shifts in settlement locations that were reportedly
responding to environmental change.
A series of observations from the Lewisville report (Ferring and Yates, 1998) and
Ray Roberts report (Ferring and Yates, 1997) shed light on prehistoric settlement
patterns and provide informative data regarding site selection along central portion of
the Elm Fork of the Trinity. Site location patterns at Lake Lewisville indicate the
utilization of ecotonal resources associated with the transition between the prairies and
12
woodlands. Faunal data documented from archaeological excavations indicates that
prehistoric populations may have been focusing on hunting on the wooded edges rather
than harvesting the potential mast crops at central locations within the Cross Timbers
(Ferring and Yates, 1998). Accordingly sites are primarily situated on the eastern edge
of the Cross Timbers near the border of the Blackland Prairie.
Topographic settings are a fundamental component of site selection that is often
incorporated but not explained interpretively. For example, one could ask why the
upland and terraces settings would be preferred over floodplains. Of the 150
components in the Lewisville Lake area 49% are located in terrace settings, 49% in
upland settings and 2% in the floodplains (Ferring and Yates, 1998). With respect to
drainages 45% are located along Little Elm Creek, 27% along the Elm Fork of the
Trinity, 20% along Hickory Creek and 8% are situated along minor tributaries (Ferring
and Yates, 1998). Given that these sites are primarily located on terraces and upland
settings rather than in the floodplain may indicate that the Lake Lewisville area was
occupied during wetter climatic episodes. Artifact assemblages excavated in this area
suggest seasonal occupations so it may be that these areas were occupied during the
wetter seasons rather than wetter climates.
The overall pattern detected by Ferring and Yates (1998: 153) suggests that
prehistoric populations “exploited whatever was available to them, and possibly did so
within quite well-defined territories. Life was rigorous and large sedentary communities
were probably never common if they existed at all.” Ferring (1998: 153) concludes that
while evidence supports an increase in resource availability, stimulated by climate
13
change, in the Late Prehistoric, there is little evidence to support that “this stabilized
culture groups as indicated by decreased mobility, increased specialization of patch
exploitation, or decreased diversity of floral-faunal resource procurement practices.”
Despite all of the investigations that have been conducted in north-central Texas it
is the consensus of several archaeologists (Lynott, 1981; Skinner and Baird, 1985;
McCormick et al., 1975; Prikryl, 1990; Ferring, 1998) that information about subsistence
and settlement patterns of the prehistoric occupants of north-central Texas needs to be
expanded upon and models refined on a regional scale. According to Ferring (1998),
the absence of a local record of past environments and chronological controls has
prevented previous settlement models from consideration of diachronic adaptive
changes. As a result, regional settlement models have not actively considered changes
in past environments as possible controls on site frequencies or site characteristics.
Throughout the discussion, of the proposed regional models, there is no mention of
sites along the West Fork of the Trinity and typically Elm Fork investigations do not
incorporate East Fork research. Although these investigators are certainly familiar with
the cultural history of the region it appears that their “regional” settlement pattern
models are situated along and primarily in reference to either the Elm Fork or the East
Fork of the Trinity. In addition, although the settlement models refer to environmental
settings, the bulk of the analyses revolve around archaeological remains. While the
results yielded from such analyses are informative, correlations with the micro-
environments for these sites are not directly addressed or synthesized on a regional
scale. Specialized ecosystems are referenced in several reports yet these areas lack
14
definition in relation to the sites and associated prehistoric behavioral patterns. In
some cases the economic utility of the environment is addressed but these data are not
directly related back to site-specific interpretations. Most investigators refer to the
general physiographic characteristics and then shift focus to describing the archaeology.
While the archaeology is of primary importance and provides evidence of prehistoric
behavior it must not be viewed in isolation. Details concerning the specific ecological
setting where this evidence was found are equally important. The dynamic relationship
that exists between the two must be preserved. As such, the inter-relationships
between the artifacts and the immediate environmental setting should be evaluated and
interpreted accordingly. The current research is designed to reveal patterns in these
inter-relationships. In addition, relationships between site locations are evaluated with
respect to the multi-facetted context in which these sites are found, in order to consider
change through time, site density, occupation frequency and explain site location.
Objectives
All of the research efforts conducted in this region have provided important
contributions and valuable insight regarding prehistoric populations that inhabited
north-central Texas. However, when sites are viewed in isolation, regional settlement
pattern behavior is difficult to discern. Given the dynamic relationship of the parts
paired with the bulk of archaeological data, formulating a regional model is a formidable
task. Fortunately, geographic information systems (GIS) technology provides
organizational mechanisms to manage the wealth of data and create links between the
various components of a complex ecosystem and archaeological sites. Rather than
15
accepting the previous generalizations, concerning settlement pattern behavior in north-
central Texas, based on a few excavated sites or surveys, the objectives of this
research are to:
1. Develop research methods and integrative techniques that enhance evaluation, analysis and interpretation of regional settlement patterns.
2. Conduct analysis of a large collection of prehistoric surface sites using
excavated sites, as a means of temporal control, in order to evaluate settlement models formulated for this region.
3. Transform environmental data layers into interpretive maps for spatial and
statistical analysis.
4. Employ GIS to define site catchments, reconstruct the economic resource space around sites, extract descriptive information and attach meaning to these attributes for interpretive analysis.
5. Test a series of hypotheses with spatial and statistical techniques to address
unanswered questions and evaluate settlement pattern theories proposed by previous investigations.
6. Provide more specificity for the interpretation of regional settlement patterns
by utilizing GIS techniques to manage large data sets, identify the particular variables that may have attracted prehistoric populations to sites and investigate adaptive changes through time.
7. Develop a methodology for future investigations and a model to explain
regional prehistoric settlement pattern behavior, in north-central Texas, that can be utilized beyond the study area, in other regions, to improve cultural resource management practices.
8. Create environmental archaeology maps to facilitate future cultural resource
management, planning, mitigation, preservation, excavation, and analysis.
Primarily, this research is conducted to shed light on prehistoric socio-economic
behavior by constructing a regional environmental archaeology database and
performing SCA to evaluate previous investigations in an interpretable environmental
setting. Focus is upon decision-making processes, related to site selection, as reflected
16
by the variability and interpretive value of soils, vegetation, faunal remains and
artifacts. To meet the objectives of this research, GIS software is utilized to place
archaeological sites in their spatial context, establish economic territories, identify
associated environmental variables, infuse the environment with interpretive meaning,
correlate artifacts and statistically evaluate a series of hypotheses. These data are
interpreted and results are synthesized in a regional model of prehistoric settlement
patterns in north-central Texas.
17
CHAPTER 2
METHODOLOGICAL FOUNDATIONS
Essentially, settlement patterns are correlates of prehistory (the archaeological
record) and geography (spatial information). Prehistory and geography are closely
linked with respect to the fact that both examine the distribution of human phenomena,
use of natural resources and inter-relations of humans with their environment (Renfrew,
1969: 74). Site catchment analysis (SCA) provides a means to evaluate these
relationships and cultural ecology, a means to interpret the associated behaviors.
Cultural Ecology
Interpretation of prehistoric behaviors is interwoven with cultural ecology. As a
science, ecology is concerned with investigating the reciprocal relationships between
living populations and the natural and social environment in which they exist (Geier,
1974: 47). Cultural ecology is “adaptation to environment” (Steward, 1955: 30). The
differentiating factor between ecology and cultural ecology is related to the added
dimension of cultural behavior which forms a dynamic part of the ecosystem. According
to Geier, this approach is more than studying relationships between man and nature,
It is concerned with the patterned interrelationships of individuals within human populations, the manner in which a population interacts with each other and with different living species; and it is concerned with the effect that change in the physical environment has on influencing and restructuring the relations of individuals and the relations between human and non-human populations in an eco-community (Geier, 1974: 47).
Since the time of Darwin, the environment has been conceptualized as the “total web of
life wherein all plant and animal species interact with one another and with physical
features in a particular unit of territory” (Steward, 1955: 30). Understanding the
18
mechanisms of cultural development begins with variation. This variation is what
defines the unique character of individual cultures. More important then the variation
itself is the source of the variation and that it can be correlated with psychological
preferences or ecological adaptations. It is this variability that explains behavioral
change and this is precisely what archaeologists seek to do.
We will never learn what causes variability in the archaeological record by simply providing a behavioral “interpretation” of our sites or if we are fortunate to have chronological data, “by reconstructing culture histories.” We must strive to learn what factors have conditioned cultural variability in the recent as well as in the distant past. This goal requires that we embrace research strategies that permit us for purposes of learning to use our prior knowledge analytically rather than simply apply it to the archaeological record in a accommodative piecemeal fashion (Binford, 2002: 472).
SCA provides a means to evaluate the ecological factors that have influenced cultural
behavior. According to Jochim (1979: 84), “until we understand the mechanisms
involved, we are limited to description and analogy.” Geier aptly points out that the
“description of a behavior pattern does not tell us why it exists only that it exists”
(Geier, 1974: 46). Cultural history is not fully interpretable without the incorporation of
the ecological framework into research strategies,
For archaeological interpretation to proceed past mere description and notation of observed patterns and regularities in material debris, the frame of reference must be extended beyond the artifact to site context, or beyond the site to the social system, eco-community, and eco-system in which it occurred. To describe a behavioral event is only one step; to understand it is the real goal (Geier, 1974: 47).
An ecological “frame of reference” (Geier, 1974; Binford, 2002) can help understand a
portion of the mechanisms involved that influence variable prehistoric behavioral
choices in relation to site selection. Developing an ecological understanding involves
19
“exploration of the resources used in relation to those available, the constraints on the
organization of procurement, the spatio-temporal variability of the environment, and
the environmental effects of exploitation” (Jochim, 1979: 89). By employing an
“ecological frame of reference”, human culture is seen as an integral part of the
environment or ecosystem, influencing and being influenced by, the patterned
interaction of its parts (Geier, 1974: 47).
Ecosystem complexity has prompted trends of ecological research in archaeology
that require multi-scalar research. Jochim (1979) cites several examples of
archaeological attempts to reconstruct prehistoric economies that correspond with this
trend. According to Jochim (1979), it is beneficial to limit the number of interrelated
variables in order to examine their interaction and reaction to change. These allegorical
windows provide brief glimpses at isolated parts of the archaeological record. In some
respects, it is necessary to break the system down into smaller components in order to
understand the parts that create the whole. However, although, the information
derived from specialized studies is of great value, it must be viewed as part of the
larger system. Focus on an isolated part causes distortion of the larger picture whereas
focusing solely on the larger picture obscures important descriptive and interpretive
details. As such, a healthy balance between these two perceptual viewpoints is
desirable. For settlement pattern studies to be potent, empirical details along with an
understanding of their cumulative effects must be interpreted, translated into
subsystems and evaluated in relation to other subsystems that form patterns within the
larger system.
20
A common problem inherent in utilizing a theory of cultural ecology relates to the
concept of “environmental determinism.” According to Steward (1938), incorporating
ecology is not environmental determinism or economic determinism. It is a
“methodological approach” to gain a deeper understanding of behavior patterns
required to survive in a given environment. Steward (1938: 261) views it as an
“equation of culture process involving the interaction and mutual adaptation of both
historically and environmentally determined behavior” rather than “a claim a priori that
ecology predetermines cultural behavior.” Even in the sciences the realization that
“physical scientific laws are not deterministic but only statistical approximations of very
high probability based on finite populations” has been accepted (Haggett, 1966: 26).
This is a key point in support of the current research. Concern is with probability
patterns that emerge from the landscape, in association with the archaeological record,
as a basis for the interpretation of regional settlement patterns. The effect of the
environmental setting on the range of possibilities available for the manifestation of
human behavior is what is appealing about an ecological approach. Even though
different economies, each composed of a system of activities that include the utilization
of different resources, may be employed within a specific environment there are still
parameters that set limitations. Any adaptation may vary only within the limits of the
character of the natural environment or the populations cannot survive. Establishment
of these limits is an important part of interpreting prehistoric behaviors and spatial
patterning. Combining cultural ecology precepts and geographical theory is the basis for
establishing these parameters and understanding the associated dynamic relationships.
21
Geographical Theory
In order to examine the spatial patterning of past human activities archaeology has
borrowed a number of analytical techniques from geography and adapted them
successfully. Locational theory, “a product of economics, ecology, systems theory, and
regional analysis” (Haggett, 1966: 26), has been applied to prehistory with little change
from its use in human geography. The text, Locational Analysis in Human Geography
served as a “major contribution to prehistoric archaeological theory” (Renfrew, 1969:
74). Central place theory first proposed by Von Thunen, in the early 1800s, developed
into a method of geographical analysis that became central to locational theory.
Von Thunen hypothesized that an isolated population center, in a uniform
environment, with an ideal distribution of production would create distinct land-use
patterns that form concentric circles radiating from the center outward. Essentially, the
further from the base site resources are, the greater their economic cost. Eventually
there is a point where cost surpasses the return. This threshold defines an economic
boundary that forms the exploitation territory of a site. A similar view developed by
Christaller (1966) incorporates site hierarchies and their related economic spheres. The
range of goods and resources available to a particular settlement are the foundation of
the hierarchies. These theories of settlement hierarchy, referred to as “central place
theory”, proposed ways in which a settlement system could be spatially organized to
perform certain types of work most efficiently (Christaller, 1966).
The location-allocation model developed by Bell and Church (1980) was formulated
to evaluate archaeological settlement patterns by assigning “relative weights to
22
strategic criteria, resource constraints, principles of economic efficiency, political
control, and transportation interconnection” (Butzer, 1982: 222). The assumption is
that settlement patterns are the “result of political, economic, and ecological forces
working themselves out on the landscape” (Butzer, 1982: 222). Jochim (1976)
developed a gravity model to analyze the interactions between human population and
several preferred resources. It operates on the premise that site locations are situated
near high density resources associated with decreased mobility (Jochim, 1976). Butzer's
(1982: 223) ecologically based resource concentration models offer effective
explanatory mechanisms that incorporate “differential plant productivity and animal
biomass of different biomes.” Such models “imply higher group density in preferred
biomes, creating stepped population gradients between marginal and optimal
environments, with major discontinuities at or near the boundaries” (Butzer, 1982:
223). These resource concentration models are viable mechanisms for forming
hypotheses and developing theories about the dynamics of hunter-gatherer behavior.
Numerous applications, of geographical theory, to archaeological problems have
been investigated. These geographically based models provide multiple perspectives
and means to observe behavior in a spatial context. Human organization across the
landscape ultimately reflects a strategic approach to survival. Geography provides
effective tools and perspectives for archaeologists to evaluate prehistoric settlement
patterns. Vita-Finzi, a geographer, and Higgs, a prehistorian, realized the benefits of
collaboration between the two disciplines when they employed a method of locational
analysis, which resulted in the development of the concept of SCA.
23
Site Catchment Analysis
Geographical theory and cultural ecology collectively form the foundation for SCA.
Central place theory forms the primary theoretical geographical foundation. Cultural
ecology includes fundamental procedures that are integral to SCA. An analysis of the
interrelationship of material culture and environment is necessary. Behavior patterns
involved in the exploitation of a particular area by means of a particular technology
must be analyzed. Environmental variables include: geology, topography, soils,
hydrology, vegetation, climate and fauna. Some subsistence patterns “impose very
narrow limits while others allow considerable latitude” (Jochim, 1979: 40). In addition,
the extent to which behavior patterns involved in exploiting a particular environment,
affects other aspects of culture, must be determined. This aspect is much more
abstract and can only be inferred based on the other two inter-relationships. Human
behavior in its various forms of interaction requires study areas of different sizes.
Three categories formulated during previous investigations include: “those focusing on
the site and its catchment, the larger sustaining region for a group and broader areas
containing several groups or larger societal units” (Jochim, 1979: 88). The current
research employs a multi-scalar approach. SCA is performed on 150 sites. These sites
are spatially and temporally analyzed to formulate explanations for regional settlement
patterns.
SCA is a method that has been utilized by archaeologists to correlate prehistoric
cultural remains with reconstructed environments (Vita-Finzi and Higgs, 1970; Schiffer,
1976; Butzer, 1982). Tiffany and Abbott (1982: 314) define SCA as an “inventory of
24
artifactual and non-artifactual remains in relation to their sources.” Its utility derives
from the “incorporation of techniques relating site location to resource availability within
a seasonally variable territory immediately surrounding a site” (Roper, 1979: 135). It
assumes that this territory is of primary importance in structuring settlement patterns.
This type of analysis provides an exploratory and explanatory mechanism for the
organization of settlements upon the landscape.
The concept of site catchment developed in response to an expressed need for
integration between environmental and archaeological aspects of the record (Vita-Finzi
and Higgs, 1970). The term “site catchment analysis” was first used in archaeological
literature by Vita-Finzi and Higgs to describe “the study of relationships between
technology and those natural resources lying within economic range of individual sites”
(Vita-Finzi and Higgs, 1970: 5). In other words, it is a technique that has been
employed to analyze the locations of archaeological sites with respect to the economic
resources available to them. SCA was derived from the movement-minimization
concept, first introduced by central place theory (von Thunen, 1966), that land use is
determined by transportation costs. SCA is typically performed by “delineating an
arbitrary economic territory surrounding a site and evaluating the resource potential
within that area” (Tiffany and Abbott, 1982: 314). Simplistically, the exploitation
territory or catchment of a site can be approximated as a circular area centered on the
site in question. The content of the catchment is interpreted by “comparative
knowledge of site territories, resource distribution, site contents and settlement system
structure” (Higgs and Vita-Finzi, 1972: 27).
25
Depending on the location, climate and time period, the environment offers very
different possibilities. Typically, people are willing to travel only so far to utilize these
resources (Roper, 1979). Sites frequently occupy positions that are not typical of their
general zonal setting.
Man sought the optimum and hence the atypical situation to meet his needs. Even if sites should prove after study to occur in an environmentally typical situation, it cannot be safely assumed at the outset that their presence is due to the factors that determine the environmental type. One means of ensuring that the environmental attributes investigated relate primarily to the site, its occupation and its occupants is by the intensive study of the small area immediately adjacent to it (Vita-Finzi and Higgs, 1970: 62).
Although this view provides information about the immediate vicinity of the site it is
important to note that exploitation territories of prehistoric sites extend beyond the
immediate vicinity especially in relation to a mobile economy.
Limitations are inherent in zonal approaches in terms of generalization but also in
catchments due to the increased mobility of hunter-gatherer populations. The
originators of the concept recognized these limitations. A human group that exploits
more than one site territory during a year may vary the intensity and focus of its
activities at certain sites. As such, it is the annual territory, rather than an individual
site territory, that accurately reflects the economic system (Jarman, 1972; Yellen,
1977). When micro-environmental settings are investigated it is important to keep in
mind the actual sphere of influence, of the particular population represented at the site,
which likely extends beyond the area (Vita-Finzi and Higgs, 1970). Therefore, a multi-
scalar approach that involves within-site and between-site investigations is necessary.
26
It is first essential to identify the territory in the proximity of a site, where primarily
subsistence needs would be met, and then expand outward. Variable techniques have
been employed to define catchment areas. Distance is one of the core variables in that
it is the most fundamental property of spatial data. Estimates of where the boundaries
of these territories should be placed have been derived from ethnography and differ
depending on the cultural economic base and landscape character. Studies of modern
hunting and gathering economies have indicated that the territory exploited from a site
tends to be within certain well-defined limits. “Range of possible behavior is both in
theory and practice limited by the capacity of a human population to exploit an area
effectively, and is further reduced once the concept of optimum is introduced” (Jarman
et al., 1970: 62). Other things being equal the further the resources are from the site
the less likely that they will be utilized. Depending on the technology available at the
time, beyond a certain distance, the area becomes uneconomical and is unlikely to be
exploited (Chisolm, 1968). Ethnographic studies indicate that the base site will be
moved, once the economic threshold has been met, to increase efficient utilization of
resources (Lee, 1968). It is important to keep in mind that these prehistoric
populations traveled by foot. “For optimal accuracy, distance should be qualified in
terms of local topography; the operative factor is the time and effort involved in
traveling the distance, rather than the absolute distance itself” (Jarman, et al., 1970:
63). Recognition of the inter-relatedness of these factors paired with the minimization
of work as a motivating factor in human behavior is the basis for constructing site
catchment parameters.
27
Studies of contemporary communities verify the importance of time, effort and
distance in defining site territories (Lee, 1968). Chisholm (1968) and Vita-Finzi and
Higgs (1970) both create an area that can be reached within 2 hours walking time, that
takes into consideration the effects of topography and accessibility, for hunting and
gathering economies. This revision of Von Thunen’s model uses measurements of
travel time rather than linear distance. Energy expenditure is considered to be more
crucial than distance in determining the limits of exploitation territories. Two-hour radii
are established on the grounds that daylight hours remaining after traveling to the
outer threshold and back would be sufficient for food preparation. Resulting territories
were compared with ethnographic case studies and found to fit fairly well with the size
of the territories exploited from home bases respectively by collecting communities.
A combination of studies by Chisholm and Lee produce similar conclusions
concerning the relationship of production to distance. These conclusions are that
“available resources become uneconomic beyond a distance of 10 km” (Chisholm, 1968:
7). Site exploitation territory, for the modern hunter-gatherer !Kung Bushmen culture,
is within a radius of about 10 km. This 10 km radius is “consistent in many parts of the
world despite technological and environmental differences” (Vita-Finzi, 1978: 26). In
Hastorf’s analysis, the calculation of site catchment is based on the maximum radius
beyond which a particular strategy is not effectively pursued. This study uses a 4km
radius for everything except nut gathering and large game hunting. “Large game
hunting and nut gathering encompassed a 10 km radius” (Hastorf, 1980: 90). Most
studies have used either circular territories of fixed radii or time contours. Particularly
28
in the case of flat or relatively uniform environments territories tend to be circular. A
variety of site catchment models have modified the parameters of the economic
boundaries, in different ways, based on the character of their region.
Site Catchment Analysis Case Studies
SCA studies have been conducted by various archaeologists over the past thirty
years to analyze the environmental context of sites, determine site function and explain
site distribution (Vita-Finzi and Higgs, 1970; Webley, 1972; Schiffer, 1976; Zarkys,
1976; Ferring et al., 1978; Flannery, 1976; Roper, 1979; Tiffany and Abott, 1982;
Butzer, 1982; Bernick, 1983). SCA has also been utilized as a means to determine the
“feasibility of various forms of economy, modeling settlement patterns and the study of
demographic processes” (Roper, 1979: 135). Inferences regarding site function have
proven useful in comparing the locations of sites. Contemporary site territories may
have resources or properties in common, or may be economically complementary (Vita-
Finzi and Higgs, 1970). Essentially, SCA focuses on the ecological, economic and
procreative aspects of human behavior that extend beyond the archaeological remains
(Hunt, 1983).
Vita-Finzi and Higgs (1970) utilized SCA techniques to establish the economic
boundaries of sites in Palestine and systematically study the resources in this area.
They classified the land within these boundaries into categories of prehistoric land use
in order to make predictions concerning resource use. Land use classifications that are
established for the current research include edibility, medicinal value, supplemental and
overall economic categories. According to these authors, some site locations are not
29
chosen primarily because of their land potential. When site locations do not correlate
with areas that have a high economic potential other factors are considered to look for
explanations for site selection. A similar approach is utilized in the current research.
Sites that lack explanation in isolation can acquire meaning when compared with other
sites (Vita-Finzi and Higgs, 1970). In addition, the character of the landscape combined
with the climatic conditions that were prevalent at the time provides clues. By
conducting intra-site comparison it has been shown that two sites situated 30 miles
apart were complementary to each other in that they provided suitable seasonal
settings for a hunter gatherer economy (Vita-Finzi and Higgs, 1970). Based on the
character of prehistoric site distribution in north-central Texas it is expected that
patterns similar to these occur in this region. SCA has also revealed that two different
cultures, as defined by their artifacts, may represent two different economies rather
than two different groups (Vita-Finzi and Higgs, 1970). Sites that are a part of the
current research are likely to contain similar artifacts. Therefore it is expected that the
economies offered by particular locations on the landscape are the primary source of
differentiation between site locations.
Environmental diversity indices for site catchments have been utilized in previous
investigations to measure the potential of an area for the presence of archaeological
sites and to determine the extent and intensity of human use (Tiffany and Abbott,
1982). The basic assumption is that site presence is correlated with increased
environmental diversity. The idea of establishing an environmental diversity index is
based on the premise that,
30
the biological and cultural needs of a society are easier to fulfill in a natural environment that has the greatest environmental diversity per area of catchment for that society. The occurrence of an archaeological site on a given portion of the landscape should be a function of the diversity of the local natural environment and likewise the greater the potential diversity of a particular study area, the more extensive and complex will be its use by groups (Tiffany and Abbott, 1982: 315).
Tiffany and Abbott (1982) consider reoccupation of a site as a function of the stability
of the local environmental diversity through time. The environmental diversity index
employed by these authors involves counting the diversity of mineralogical, hydrological
and vegetational resources within the site catchment territory. The current research
utilizes a diversity index to test for this however it is derived by different means.
Hunt conducts a case study in Northeastern America by utilizing a SCA methodology
that includes Geographic Information Systems (GIS) based analysis (Hunt, 1982). Hunt
is concerned with the landscape that surrounds the archaeological site and the concept
of site catchment as it applies to quantifying the physiographic space within and around
a site (Hunt, 1982). Hunt demonstrates the ability to characterize soil matrix around
Late Woodland village sites in a multitude of ways and to link soil data to other
attributes (Hunt, 1982). Webley’s (1972) analyses of Tell Gezer and several other sites
in Palestine are based entirely on analysis of soils and their potential productivity
(Roper, 1979: 126). While such studies are limited to description of how sites are
located relative to soils or vegetation, soils analysis offers valuable information to
cultural history, including comparisons among sites of their potential for certain
economic activities. According to Roper (1979), focusing on soils or vegetation in
isolation may not explain why a site is located where it is or how it functioned in a
31
settlement system. More complete modeling of settlement location and the settlement
system requires the use of a wider variety of resource types (Roper, 1979). While the
current research is highly dependent on soils, to economically interpret site catchments,
other factors are taken into account to provide explanatory mechanisms. Each
component is a piece of the puzzle that fits together to form a more comprehensive
view of prehistoric behavior.
In a study about subsistence strategies of prehistoric populations that inhabited the
Mimbres River Valley in New Mexico, SCA is the principal technique used to collect and
systemize information (Hastorf, 1980). In this study conducted by Hastorf (1980) six
food gathering strategies are described, and by SCA, potential yields for each strategy,
in each time period, are determined. The economic model provides the framework
within which the strategies and their potential yields are organized and graphically
depicted. According to the model, “in a subsistence economy, population size
determines both caloric requirements and marginal costs needed to meet those
requirements” (Hastorf, 1980:80). The assumption is that there is a definable area for
each food-procuring strategy associated with each settlement (Hastorf, 1980). SCA
translates environmental information into potentiality for particular strategies. SCA is
used to determine a land-use area for each strategy for each time period. Once this
area is defined and the caloric productivity per unit is calculated, the potential yield for
each strategy is obtained (Hastorf, 1980). Given the character of the prehistoric sites in
north-central Texas this type of approach is not directly applicable for the current
research.
32
Presence of the unique floral and faunal material excavated at a site on Vancouver
Island, in British Columbia, prompted a site catchment analysis (Bernick, 1982).
Instead of starting with the site territory and reconstructing the economic patterns, an
analysis conducted by Bernick (1982) begins with the set of excavated cultural remains
that imply particular subsistence strategies. The exploitation territory is delineated
based on these data. Her objective is to define the spatial dimensions of economic
activities and to assess the applicability of the site catchment model to analogous sites
on the Northwest Coast (Bernick, 1982). She factors in a defined travel time required
to procure specific resources and the topographical location of sites. Confirmation of
hypotheses is interpreted as an indication that the site catchment model is effective for
evaluation of economic behavioral patterns of Northwest Coastal prehistoric
communities (Bernick, 1982). This case study is an excellent solution for site specific
investigations. However, on a regional scale, currently the data are lacking to
successfully carry out a methodology such as this one.
Statistical techniques are an effective mechanism for objectively evaluating site
catchments in relation to their immediate resource potentials. Previous SCA studies
have utilized statistical techniques to accomplish this goal. Multivariate statistical
techniques such as “factor analysis, multidimensional scaling and cluster analysis” are
utilized by Roper for description and comparison of site territories with their resource
potential (Roper, 1979: 127). Zarkys (1976) analysis of site catchments at Ocos in
Guatemala used “percentage point differences, chi-squared tests, and binomial tests for
evaluating which environmental zones were represented in higher proportions
33
immediately surrounding a site than they were in the total study area” (Roper,
1979:128). Kvamme and others followed a similar methodology comparing site
selection to overall regional characteristics incorporating a GIS to facilitate analysis,
A raster GIS, together with quantitative analysis software by the SAS institute Inc. [1985], provides an ideal tool for spatial analysis because a variety of archaeological and biophysical data can be co-registered to vast numbers of locations cross broad regions, allowing rapid investigation of relationships between archaeological features and any number of locations across broad regions, allowing rapid investigation of relationships between archaeological features and any number of environmental parameters (Kvamme, 1989: 168).
A variety of statistical approaches have been employed to condense vast amounts of
data into interpretable statistics. However, it is important to note sample size and
archaeological recovery bias. Particularly, in the current research, a strong bias is
already present in the amount of contract work that has been conducted along the Elm
Fork of the Trinity due to reservoir construction. Considerably less work has been
conducted in Collin County and even less in Wise County. As a result, the current
statistical results are heavily influenced by recovery bias.
SCA has proven to be a very useful methodological technique that has provided
answers to a plethora of questions to meet the objectives of a variety of research
efforts. Ultimately, SCA provides a means to develop a framework for investigating the
distribution of resources in relation to site location. The rationale for SCA operates on
the assumption that,
human beings are refuging animals, rhythmically dispersing from and returning to a central place (Hamilton and Watt, 1970: 263), differentially using a seasonally and spatially variable landscape in a manner that generally is conservative of energy, but conservative relative to a relative scale of values placed on needs and wants (Roper, 1979:122).
34
Given that such behavior takes place in particular locations it is logical to extract and
evaluate environmental attributes from catchments of archaeological sites to shed light
on prehistoric behavior. The theoretical foundation of the SCA model is a sound and
viable basis for the current research. This method is a useful way to integrate the
archaeological investigations conducted in the region and a means to systematically
study sites in relation to the dynamic character of the ecosystem.
Although the current research employs SCA techniques that are similar to those that
have been utilized in previous investigations it does not follow any of these research
methodologies in particular. Given the data limitations and character of the region, the
current research design incorporates a variety of techniques suggested by several
studies, modifies them and adapts them to meet the needs of the current research.
This methodology is designed to provide answers to the unanswered questions related
to prior research conducted in the region and formulate a model of prehistoric
settlement pattern behavior.
Geographically, factors such as topography, geology, soils, vegetation, water
sources, fauna and land suitability influence past cultural behavior. Identifying the
specific cultural and non-cultural characteristics that factored into site selection with
respect to site catchment variability is pertinent to understanding the dynamic socio-
economic prehistoric behavioral patterns in north-central Texas. In order to gain a
deeper understanding of these patterns it is appropriate to begin with an evaluation of
the natural setting (environment) followed by the synopsis of the defining
characteristics of the various populations (cultures) that inhabited the region.
35
CHAPTER 3
RESEARCH SETTING
North-central Texas is located in an advantageous position within the “bioclimatic
transition between the forested Gulf Coastal Plain and the rolling Southern Plains”
(Ferring, 1997:19). Increased “biological diversity” is the result of numerous factors,
including the region's geologic, soil and climatic variation combined with its eco-tonal
location between the “deciduous forests of East Texas and the grasslands of the
Southern Plains” (Diggs et al., 1999: 20). The combination of these factors provides a
highly productive blending ground for plants and animals from the east and west. In
addition, the increased diversity of this region offers a wide range of economic
opportunities for prehistoric populations.
Geology of North-Central Texas Geological structures in north-central Texas were shaped by uplift and formation of
an extensive mountain system that includes the “Appalachians, Wichitas and Ouachitas”
(Diggs et al., 1999: 21). Through time, shallow seas repeatedly advanced and
retreated over most of Texas depositing a number of different layers of Cretaceous
sediments (Diggs et al., 1999). As a result of these depositional processes and
subsequent erosion, surface rocks in north-central Texas are predominantly Cretaceous
in origin. Numerous layers of limestone, marl, shale, and sandstone were deposited
over the area. Since climatic variation in north-central Texas is minor, differences in
landforms, soils and vegetation are attributed to bedrock lithology (figure 2) (Ferring
and Yates, 1998: 5). Lynott (1981) describes the geology as, “dominated by a series of
36
eastward dipping sedimentary marine deposits.” Pennsylvanian deposits are exposed in
the western parts of the region, while Cretaceous clays and sands dominate the eastern
prairies. These eastward dipping strata (figure 2) consist of Antlers Sandstone, Denton
Clay, Woodbine Sandstone, Eagle Ford Shale, Austin Chalk and Ozan Marl interspersed
by resistant outcrops known as the Goodland, Kiamichi, Pawpaw, Weno, and Grayson
formations. The entire Upper Trinity drainage basin has developed over these
Cretaceous and Pennsylvanian sedimentary rocks (Ferring, 2001; Diggs et al., 1999).
North of the West Fork of the Trinity the “Antlers Formation is primarily composed of
fine-grained sandstone and shale” (Ferring and Yates, 1997: 19). South of the West
Fork the “Twin Mountains, Glen Rose and Paluxy Formations have a more diverse
lithology” (Ferring and Yates, 1997: 19). The upper part of the West Fork drainage,
northwest of Fort Worth, formed on top of Pennsylvanian limestone, shale and
sandstone. All other portions of the Upper Trinity drainage basin developed over
Cretaceous rocks (Ferring, 2001). Geology is culturally significant in that it provides
resources for tool manufacture and settings for sites. In addition, the characteristics of
the geology influence the character and range of potential soil types that can develop.
Geology, combined with the climate, organisms, and topography ultimately determine
the range of possible vegetation that can be produced in a given area. In north-
central Texas limestone settings have predominantly produced prairie vegetation and
sandstone settings have provided the foundation for forest vegetation. The variation
within these geological formations is illustrated (figure 2) from a bird’s eye view.
37
Figure 2. Geology of north-central Texas (based on USGS; Ferring and Yates, 1997).
38
Topography
The topographic expression of geological formations in north-central Texas
originated from tectonic activity, erosion and variable depositional processes. Elevation
in north-central Texas generally decreases from west to east. Topography is nearly
level to very hilly. Wise County is transected by the Fort Worth Basin which combined
with the uplifted resistant bedrock creates a highly variable landscape with steep
slopes. The elevation reflects this variability ranging from 1,286 feet in the southwest
to 649 feet in the eastern portion of the County. Denton County, although also
transected by the Fort Worth Basin in the southern portion of the county, is less
variable with very few resistant outcrops. This is reflected by elevations that range
from 900 to 500 feet. Elevations, in Collin County, reflect the less resistant limestones,
shales and marls, range from 900 to 400 feet, gradually decreasing eastward. Collin
County is situated in an area where the Fort Worth Basin and East Texas Basin merge.
This topographic setting reflects relatively lower elevations that explain the increased
number of sites associated with bottomland settings along the East Fork of the Trinity.
Topography in north-central Texas combined with the character of the geology
ultimately influences and is influenced by the character of the drainage networks.
Drainage Systems
Drainage systems have actively sculpted the landscape forming as a result of the
topography and resistivity of the geology. Northern Texas is drained by a network of
rivers that includes the Brazos, Canadian, Colorado, Red River, Sabine, Sulphur and
Trinity (figure 3). One of the most dominant features transecting the landscape of
39
F
igure 3. Hydrological networks of north-central Texas (based on TNRIS data).
40
north-central Texas is the Trinity River. From its beginning, northwest of Fort Worth in
Archer County, the Trinity flows to the southeast approximately 692 River miles to the
Gulf Coast with a change in elevation of 1,250 feet (Nixon and Willett, 1974). The
Upper Trinity drainage basin is surrounded by three major drainage basins: the Red
River to the north, the Brazos River to the southwest, and Sabine River to the east.
The Trinity has been a focal point of cultural activity for thousands of years.
The West Fork of the Trinity flows in a southeasterly direction through Jack, Wise,
Tarrant and Dallas Counties until it forms confluence with the Elm Fork of the Trinity.
The West Fork is a consequent stream of the Upper Trinity drainage basin that has
incised into the resistant Woodbine sandstone and Austin chalk (Ferring, 1998).
Although archaeological sites are rarely reported, along the West Fork, within the
confines of Wise County, there is considerable undocumented evidence to suggest that
prehistoric populations aggregated along this major tributary. Denton Creek is a major
tributary in the Upper Trinity Basin that forms a part of the Trinity “fork” system. It is
referenced due to its archaeological significance and proximity to the West Fork. The
headwaters of Denton Creek begin in Montague County and flow to the southeast
through the northeast corner of Wise County and the southwest corner of Denton
County where it veers east to form a confluence with the Elm Fork. For the current
research sites near Denton Creek will be included in the West Fork study area
The headwaters of the Elm Fork of the Trinity are in the west central section of
Montague County near the Cooke County line. From this point it flows in an easterly
direction to central Cooke County where the principal drainage curves to the south and
41
then southeast. This general course is maintained to the stream's confluence with the
West Fork of the Trinity near Dallas. The vast majority of sites in this region are
associated with the Elm Fork of the Trinity.
The headwaters of the East Fork of the Trinity originate in Grayson County flowing
south through Collin, Rockwall, and Grayson Counties until Ellis County where it forms a
confluence with the Trinity. This major tributary is situated in a unique environment
centrally located in the Blackland Prairie where it serves as the only major intermediary
water source between the main stem of the Trinity and the Sabine River.
All of these drainages follow the tilt of the regional bedrock forming elongated
dendritic patterns. The West Fork, Denton Creek, Elm Fork and East Fork all merge into
one by the time the River system reaches Ellis County where the Trinity guides the flow
to the Gulf of Mexico. In prehistoric times drainage systems acted like highways linking
separate geographical areas together. In addition to being culturally significant, these
hydrological networks are responsible for the variable character of the landscape.
Physiographic Setting
Four major upland physiographic subdivisions (figure 4) have been recognized in
north-central Texas: the Western Cross Timbers, Grand Prairie/Fort Worth Prairie,
Eastern Cross Timbers and Blackland Prairie (Dyksterhuis, 1948; Ferring and Yates
1997). The character of the four major upland geomorphic subdivisions is influenced
mostly by the bedrock lithology and the correlative soils. These alternating strips of
prairies and oak woodlands form a landscape that is mosaic in character and has
significantly contributed to the increased diversity of north-central Texas.
42
Figure 4. Physiographic regions of Texas and north-central Texas (TNRIS).
43
The Western Cross Timbers is a physiographic subdivision that occurs in the western
part of the Upper Trinity Basin. This region, situated mainly in the Wise County portion
of the study area, is a rolling to deeply dissected area with sandy soils that corresponds
with the Antlers Formation (Ferring and Yates 1997; Ferring 2001). It is characterized
by a rolling to hilly topography with a very sandy soil cover. The Western Cross
Timbers are similar to the Eastern Cross Timbers, yet less rugged in character due to
the corresponding geology.
The Grand Prairie physiographic subdivision occupies the central part of Denton
County and the eastern portion of Wise County. Alternating beds of shales and
limestones stratigraphically situated between sandstone formations provide the
geological foundation. Typically, it is a “rolling upland prairie with small escarpments
and benches” formed by geological layers such as the Goodland formation, Duck Creek
limestone, Fort Worth limestone and Main Street limestone (Skinner, 1973: I-II). The
Fort Worth Prairie is a level to gently rolling surface, nested in the central portion of the
Grand Prairie, north of the West Fork of the Trinity. The foundation of this area is a
resistant Cretaceous limestone, composed of the Goodland Formation in the west and
Grayson Formation in the east. “Differences in lithology have primarily controlled
development of the upland prairie topography and tributary alluvial valleys” (Ferring,
2001: 27).
The Eastern Cross Timbers are situated along the central and eastern boundary of
the Upper Trinity Basin. This physiographic region follows the Woodbine sandstone
forming a narrow north-south belt of low hills that rise above the prairies on either side
44
(Ferring and Yates, 1998). Due to the fact that the Woodbine formation is not as thick
as the Antlers Sandstone, the Eastern Cross Timbers are narrower than the Western
Cross Timbers (Ferring and Yates, 1998). The woodbine formation consists of
sandstones, shales, sandy shales, and fine to coarse sands providing the foundation for
the Eastern Cross Timbers. Since the sandstones are more resistant to erosion than the
clays and shales “the topography is more hilly and rugged in character” than the
Western Cross Timbers (Prikryl, 1990: 6).
The Blackland Prairie physiographic subdivision is immediately east of the Eastern
Cross Timbers. At the eastern margin of the area the “Blackland Prairie developed on
relatively young Cretaceous layers” (Diggs et al., 1999: 21). Most of the Blackland
Prairie corresponds with Upper Cretaceous rocks, including the Eagle Ford Formation
(shale, clay, and marl), the Austin Chalk and the Ozan Marl (Ferring and Yates 1997;
Ferring 2001). This physiographic region is characterized by a fairly flat to rolling
surface developed on the Eagle Ford Shale and Austin Chalk formations (Dyksterius,
1948). These patterns reflect the dominant influence that the bedrock has over the
region.
The Western and Eastern Cross Timbers, interspersed by prairie settings, form a
diverse physiographic region that is related to the geology, topography, soils and
vegetation. These physiographic settings are described in more detail in the vegetation
section following a brief review of the soils formed in north-central Texas. The
character of the soil is highly dependent on the geology and forms the basis for the
range of vegetation available for prehistoric utilization.
45
North-Central Texas Soils
Soils in north-central Texas vary dramatically, ranging from the erosive sandy soils
that form the foundation for the Western Cross Timbers to the highly plastic clay soil of
the Blackland Prairie. The character of the soil is interdependent upon the geological
parent material, topography, the organic and biotic community and the climate.
Comparisons of the soil taxonomy within the prairie and forest environments provide a
greater understanding of the range of variation and differences in vegetation that can
potentially form in each setting. West Fork soils are composed of Western Cross
Timbers, Grand Prairie and Fort Worth Prairie soils from west to east (figure 5A). Elm
Fork soils consist of Grand Prairie, Eastern Cross Timbers and Blackland Prairie soils
from west to east (figure 5B). East Fork soils are entirely composed of Blackland Prairie
soils (figure 5C). An overview of the great groups present in the study area illustrates
the soil distribution referred to in the following discussion (figure 5).
Western Cross Timbers Soils
Soils of the Western Cross Timbers are mainly Paleustalfs. These “Alfisols are
associated with the sandstone bedrock and sandy alluvium, whereas the Mollisols and
Vertisols are associated with the calcareous bedrock and alluvium” (Ferring, 2001: 27).
With the exception of the soils in the western portion of the Western Cross Timbers that
developed on gravelly and rocky Pennsylvania strata the soils of the East and Western
Cross Timbers mainly developed on sandy Cretaceous Woodbine strata. Mollisols and
Vertisols are both typically associated with the grassy areas of the Cross Timbers
whereas the Alfisols are associated with increased production of hardwoods.
46
Figure 5. Soil distribution in north-central Texas study area.
47
Duffau-Keeter-Weatherford, Windhorst-Chaney-Selden, Truce-Cona and Bastsil-Silawa
are general soil map units associated with the Western Cross Timbers (Ressel, 1989).
These loamy and sandy, well drained soils composed of material weathered from
sandstones and shales, formed upon a dominantly gently sloping surface on upland and
terrace savannahs (Ressel, 1989). Great groups associated with these units are
Paleustalfs and Haplustalfs (Ressel, 1989). A mixture of these older more developed
and younger minimally developed Alfisols maintain a fair amount of moisture with dry
intervals. Climax vegetation includes oaks, with an under story of grasses, peanuts,
grapes, tree fruits and nuts (Ressel, 1989). Woody and herbaceous vegetation provide
food and cover for deer, turkey, squirrel, dove, and quail (Ressel, 1989).
Grand Prairie and Fort Worth Prairie Soils
Soils of the Grand Prairie and Fort Worth Prairie have similar characteristics to the
Blackland Prairie but the shrink-swell properties of the clay are less intense. Differences
between the prairie environments can be attributed to the bedrock lithology. Mollisols
are primarily found on the Grand Prairie and Fort Worth Prairie on various limestone
layers (Diggs et al., 1999). “Mollisols are less problematic than Vertisols in terms of
shrink-swell (Diggs et al., 1999: 25) and are considered among the world’s most
productive soil due to their high native fertility. Soils on the Fort Worth Prairie vary
according to bedrock parent material. Most upland soils are clayey and calcareous
Chromusterts, Calciustolls or Haplustolls.
General soil units associated with the Grand Prairie include the Sanger-Purves-
Somervell and Venus-Aledo-Somervell units (Ressel, 1989). These units are clayey and
48
loamy, well-drained soils that formed in material weathered from limestones and marls
on upland prairies (Ressel, 1989). Related great groups include Chromusterts,
Calciustolls, and Haplustolls that indicate a mix of Mollisols and Vertisols in moist and
dry climates (Ressel, 1989). The presence of Mollisols and Vertisols indicate a prairie-
dominated environment. Yet cobble and limestone fragments contribute to a lithic
component that creates a habitat that is unsuitable for most wildlife (Ressel, 1989).
Aledo-Somervell soils are well-drained loamy soils formed in material weathered
from limestone in the Uplands (Ressel, 1989). These soils are not very common and
are focused in two particular locations near Denton Creek and Clear Creek. The Aledo
component is made of Lithic Haplustolls while the Somervell is Typic Calciustolls
(Ressel, 1989). Both are composed of Mollisols in fairly moist to periodic dry regimes
but the former has a stony character with minimum horizon development whereas the
latter is more calcareous indicating increased development. According to the soils
survey, 90% of the vegetation consists of varieties of grasses and the remaining portion
consists of various trees and shrubs (Ressel, 1989).
Sanger-Somervell, Navo-Wilson, Slidell-Sanger and Ponder-Lindale are fairly well-
drained clayey and loamy soils that formed in material weathered from calcareous clay
and marl in the upland prairies (Ressel, 1989). Chromusterts, Calciustolls, Paleustalfs,
Ochraqualfs, Pellusterts and Halustalfs indicate a diverse mixture of Vertisols, Mollisols
and Alfisols revealing the gradual shift from the Grand Prairie to the Eastern Cross
Timbers (Ressel, 1989). Vegetation associated with these soil units is dominated by
grasses and forbs (Ressel, 1989).
49
Eastern Cross Timbers Soils
Soils of the Eastern Cross Timbers are primarily Alfisols that developed from the
sandy woodbine formation. Alfisols are generally productive soils that promote the
growth of hardwood forest vegetation and crop yields. Principal surface soil
associations are summarized by the Birome-Gasil-Callisburg unit that contains well-
drained loamy soils formed in sandstone and clay parent material (Ford and Pauls,
1980). Great groups associated with this soil unit are Paleustalfs indicating a fairly
moist climatic regime (Ford and Pauls, 1980). Paleustalfs are the most dominant great
group in the Eastern Cross Timbers with minor variations related to the clay parent
material. On the sandy Alfisols, the woody plants--namely post oak, blackjack oak, and
hickory are dominant; while the grasses prevail on the tighter, more drought-resistant
Mollisols (Ford and Pauls, 1980). Oak is the most common tree reported in the Eastern
Cross Timbers (Ford and Pauls, 1980). River valleys crossing the region support a forest
of hackberries and pecans mixed with oaks on the alluvial soils (Ford and Pauls, 1980).
Blackland Prairie Soils
Soils in the Blackland Prairie are composed of the Houston Black-Austin group,
Houston Black-Houston, Ferris-Houston and Wilson-Burleson that consist of clayey soils
formed in marl and chalk parent material (Wheeler and Hanson, 1969). Primary
differences between these general soil units are found in the structural density of the
clay, susceptibility to erosion and slope orientation.
Three dominant soil orders identified in the Blackland Prairies include Alfisols,
Vertisols and Mollisols. Alfisols are less fertile than Mollisols or Vertisols and are found
50
mainly on the Eastern and Northern margins of the Blackland Prairie. Vertisols develop
mainly on the Eagle Ford Shale or rocks of the Taylor Group and are characterized by a
high clay content that exhibits increased shrink-swell properties. “Swelling and shrinking
causes deep and wide cracks” (Diggs et al., 1999: 23). The “black waxy” soils of the
Blackland Prairie are derived from rocks of limestone origin (Diggs et al., 1999: 23).
This “incredibly sticky soil has historically been particularly problematic and virtually
impassable during wet weather” (Diggs et al., 1999: 25). Drought however is
considered to have been more of a problem, with a lack of water considered a limiting
factor for humans (Diggs et al., 1999). Impermeable black clay soils, the lack of
dependable water bearing layers, and the ephemeral nature of surface streams made
the “early Blackland Prairie a particularly inhospitable environment for humans” (Diggs
et al., 1999: 29). Vegetation provides sustenance consisting of grasslands with a
scattered mixture of hardwoods and particularly pecans. These areas are associated
with bottomland hardwoods that provide a suitable habitat for wildlife.
Riverine and Floodplain Soils
Riparian corridors provide unique micro-environments that are significantly different
from the generalized landscapes of the physiographic regions. Similarities can be
attributed to bedrock lithology and differences to the influence of concentrated water
resources combined with the variation in the alluvial parent material. Riverine settings
provide increased productivity and tend to produce hardwoods that flourish in riverine
settings. Characteristic vegetation and faunal resources found along streams and
riverbanks reflects the variability of these settings.
51
The Pulexas-Balsora-Deleon unit is situated on the floodplains of the West Fork of
the Trinity in the Western Cross Timbers (Ressel, 1989). It consists of loamy and
clayey, moderate to well-drained soils that formed in alluvium. Associated great groups
suggest that these soils are primarily recent Ustifluvents and Mollisols that exhibit
minimum horizons (Ressel, 1989). Vegetation includes an overstory of oak, elm,
hackberry, pecan and willow varieties. Small grains, peanuts, melons and sorghum are
common (Ressel, 1989). These soils have a strong potential to support a wide variety
of wildlife.
Frio-Ovan soils are well-drained, clayey soils that formed in material weathered from
alluvium in the bottomlands that correspond with stream channels transecting the
Grand Prairie and Eastern Cross Timbers (Ford and Pauls, 1980; Ressel, 1989).
Related Haplustolls and Chromusterts indicate a mix of Mollisols with minimally
developed horizons and Vertisols with a high organic content (Ford and Pauls, 1980;
Ressel, 1989). These soils are concentrated in the floodplains of Denton Creek, Hickory
Creek, Clear Creek and the Elm Fork of the Trinity. Vegetation related to these soils is
composed of approximately 85% grasses with 15% representation of various trees such
as pecan and elm (Ford and Pauls, 1980; Ressel, 1989).
The Frio-Trinity soils are loamy and clayey, moderate to well-drained soils that
formed in alluvium on the floodplains in the Grand and Blackland Prairies (Ressel, 1989;
Wheeler and Hanson, 1969). The Frio-Trinity component contributes Haplustolls and
Pelluderts soils situated around portions of Denton, Catlett and Sweetwater Creeks.
Mollisols and Vertisols associated with these great groups indicate increased
52
productivity paired with a shrink-swell component (Ressel, 1989; Wheeler and Hanson,
1969). Soil units associated with riparian environments within the Blackland Prairie
include the clayey and loamy soils of the Trinity-Frio found on the floodplains and the
clayey soils of the Houston-Black Burleson found on stream terraces (Wheeler and
Hanson, 1969). Many native pecan trees are located along the stream channels and
interspersed on the floodplains (Ressel, 1989; Wheeler and Hanson, 1969). Yields from
Pecan harvests would have provided a very desirable resource with a high-caloric value
for prehistoric populations. The presence of this soil unit is so concentrated and
economically valuable that prehistoric presence in association with these soils would
strongly suggest a direct correlation with vegetation.
Floral Resources in North-Central Texas North-central Texas is an interesting and diverse region in terms of vegetation,
which responds to a combination of soils and climate. Following the physiographic
character of the landscape, vegetation in Texas is divided into several zones known as
the Pineywoods, Gulf Prairies and Marshes, Post Oak Savannah, Cross Timbers and the
Blackland, Fort Worth and Grand Prairies. Oak-Hickory forest and tall-grass prairies
transect this region. North Texas flora includes “2223 species (46% of the species
known for Texas) and a total of 2376 taxa (species, subspecies and varieties)” (Diggs et
al., 1999: 20). Due to increased biodiversity related to the Cross Timbers and Prairies
different microhabitats within the same county contain different plant communities.
The regional study area is composed of a combination of four distinct geographic
provinces including the Blackland Prairie, Eastern Cross Timbers, Grand Prairie, and
53
Western Cross Timbers. These zones extend, southward about 150 miles and
northward to the Arkansas River, as continuous bodies of forest or savannah bounded
by open prairie (Dyksterhuis, 1948). The Eastern and Western Cross Timbers reside
east and west of the Grand and Fort Worth Prairies. Resources associated with prairies
and forests offer a vast range of opportunities. This geographic expanse provides a
unique opportunity to investigate prehistoric behavioral choices, in relation to an
environment situated within an eco-tonal region that provided a diversity of resources.
Prairies
The predominantly clay-rich soils of the prairies support tall to mid-size grasses with
clusters of oak trees in the uplands and riparian communities that follow the waterways.
Vegetation along the waterways in the Prairies consists of similar understory species to
those found in the temperate woodlands of the Cross Timbers (Dyksterhuis, 1946;
Ferring and Yates, 1997). Climax vegetation in the Blackland Prairie is composed of
mixed grass prairies, with scattered stands of cedar and oak trees situated on the
"White Rock Escarpment" near the western edge of the Austin Chalk (Ferring, 1997:
22). Kindall, an explorer, in 1845, described the Blackland Prairie as “varied by small
prairies and clumps of woodland”, and the Fort Worth Prairie as “a perfect ocean” of
grasses (Ferring and Yates, 1998: 12; Diggs et al., 1999). Dark clay soils of the
Blackland and Grand Prairies sustain short grass prairies (Skinner and Baird, 1985: 2-2).
Dominant species are bluestem, speargrass and wheatgrass. Prairie vegetation includes
tall and midgrasses such as bluestems and gramas. Buffalo grass and species of three-
awn, among others, tend to increase under grazing. Over “250 species of plants have
54
been identified for these prairies” (McCormick et al., 1975: 39-40). Trees such as bois
d’arc, oak, elm, ash, pecan and post oak trees dot the landscape along stream channels
(McCormick et al., 1975). Brush species have invaded the prairie, and weedy annual
and perennial grasses have increased in number (Ressel, 1989). The Blackland Prairie
offers numerous seed bearing trees, plants and grasses. “Large herbivores traditionally
migrated through this area in the fall" (McCormick et al., 1975: 39-40). Streams,
transecting the prairies, provided “water and riverine resources including both
nutritional and medicinal trees and plants” (McCormick et al., 1975: 39-40).
Cross Timbers
North-central Texas contains an oak dominated vegetation region that is part of a
much larger area referred to as the Cross Timbers. Pollen data reveals that the “forests
of the Cross Timbers were not established until the middle to late Holocene” (Ferring,
2001: 28). In Texas, this region is composed of two strips, the Eastern and Western
Cross Timbers, separated by tall grasses and the Grand Prairie. Contrasts in vegetation
between Prairies and Cross Timbers create natural barriers and facilitate north-south
trails,
The conspicuous change in vegetation from the Blackland Prairie to the Eastern Cross Timbers served as a landmark on early maps (e.g. Holley, 1836; Ikin, 1941; Gregg, 1844; Kendall, 1845) and was discussed in many immigrant guides and early explorer accounts (e.g. Marryat, 1843; Roemer, 1849; Parker, 1856). Kennedy (1841) wrote, “This belt of timber varies in width from five to fifty miles. Between the Trinity and the Red Rivers it is generally from five to nine miles wide……When viewed from east or west, it appears in the distance as an immense wall of woods stretching from north to south (Diggs et al., 1999: 43).
55
The Cross Timbers have also been described as a “singular strip of wooded country with
an almost impenetrable undergrowth of brier” that was “broken” and “full of deep,
almost impassable gullies” (Ferring and Yates, 1998: 12). This thick vegetation, coupled
with the heavily dissected topography, led early travelers to comment on the difficulty
of traversing the Cross Timbers (Prikryl, 1990; Diggs et al., 1999). Early accounts of
the Cross Timbers also report of oak varieties with a dense under-story of saplings,
vines and greenbrier (Diggs et al., 1999). Of the dominant trees, post oak is the most
widespread. Hickory, which requires the most moisture, is the least common. “In the
1940s, wooded overstory covered 35% of the land surface in the Western Cross
Timbers” (Dyksterhuis, 1948: 337). This consisted of “63% post oak and 29%
blackjack oak with a remaining 8% composed of species such as hackberry and cedar
elm” (Dyksterhuis, 1948: 337). The primary difference between the Cross Timber areas
is found in the rate of precipitation that increases from west to east. Therefore, it is
anticipated that the Eastern Cross Timbers would exhibit a relatively increased
productivity. However, while there is a good variety in the subsistence resources of this
vegetation region, it is an area that is prone to droughts (Story, 1985). Western and
Eastern Cross Timbers are major areas of oak-hickory, with open savannah, dense
brush, post and blackjack oaks. Eastern Cross Timbers are essentially a mix of the
Western Cross Timbers and Oak Woodlands.
Vegetation of the Eastern Cross Timbers can be regarded as a western fringe area
of the eastern woodlands. Over three hundred species of plants are native to this area
(Tharpe, 1926). Forest vegetation is dominated by post oak, blackjack oak, pecan and
56
elm. Hickory is virtually replaced in the Eastern Cross Timbers by the Pecan tree. Main
components of the understory are little bluestem ragweed and elderberry. Small
patches of prairie within the Cross Timbers are dominated by grasses such as
wheatgrass, bluestem and speargrass. “Species diversity of dominant terrestrial plants
for the upland post oak forest and the streamside forest is low” (Skinner and Baird,
1985: 2-3). Cross Timbers areas contain sandy upland soils that support dense groves
of post oaks and greenbrier with an understory of big and little bluestem, switchgrass,
lovegrass, sumac, plum, many varieties of forbs, legumes and other shrubby vegetation
(Ferring, 1990; Story, 1985). In addition, these sandy areas often produced peanuts.
Post oak, blackjack oak and hackberry are also particularly abundant in wooded uplands
(Story, 1985; Prikryl, 1990). On the floodplains in the Cross Timbers, “tree cover
consists of elms, pecans, oaks, cottonwoods, and willows with an understory vegetation
that includes grasses, a variety of forbs and various fruit-bearing plants such as plums
and berries as supplements to a rich variety of nut-producing trees” (Ferring and Yates,
1998: 12). Species associated with the well watered stream valleys of the Cross
Timbers, include pecan, walnut, willow, elm, sycamore, and cottonwood (Story, 1985).
Along the Elm Fork of the Trinity in the Eastern Cross Timbers a mixture of ash,
blackjack oak, black willow, burr oak, elm varieties, hackberry, honey locust, mulberry,
pecan, post oak and wild plum are available (Story, 1985; Prikryl, 1990). Trees from a
variety of settings, including floodplains, uplands and stream valleys, within the Cross
Timbers provided a valuable food source, as well as, fuel for fire.
57
The Western Cross Timbers share many of the same characteristics as the Eastern
Cross Timbers separated only by the vegetation of the Grand Prairie. Stream
floodplains are dominated by various hardwood species and Juniper clings to the steep
slopes along Rivers. The natural vegetation is a savannah of oaks that follow the
stream channel or cluster across the landscape accompanied by an under-story of mid
and tall grasses. A variety of plant species create diversity with a patchy, mosaic
distribution across the landscape. Even with the wide variation in soils, the climax
understory vegetation is fairly uniform with the predominant grasses being little
bluestem, big bluestem, Indian grass, switchgrass, Canada wild-rye, side-oats grama,
hairy grama, tall dropseed and Texas wintergrass (Ressel, 1989; Prikryl, 1990).
In addition to the two main vegetation zones, Prairies and Cross Timbers, two other
additional vegetation habitats are significant in the region and are important to
understanding prehistoric settlement patterns in relation to plant resources. First, the
transitional zones between physiographic settings referred to as the ecotones. These
ecotonal settings contain plant species that are common to both physiographic settings
and there are some species that are exclusively found in these zonal interfaces. This
transitional zone effectively serves as an area where a number of species common to
woodlands and prairies would have greater accessibility. Another small, but important
habitat would be the riparian and floodplain zones. A large number of trees, shrubs,
berries and other vegetation types are located in this zone and are not found
elsewhere. This is due to the more mesic microenvironment of the riverbank and
58
floodplain in contrast to the more xeric terrace habitats. Here a number of species
uncommon or absent elsewhere would be available in a spatially limited area.
Investigations of McCormick et al. (1975) suggest that floral resources are equally
distributed over the uplands, slopes, and bottomlands. Many individual species occur in
more than one environmental setting (Prikryl, 1990: 12). For example, roots, such as
prairie turnips, were probably an important food source available in both the post oak
savannas and prairies (Martin and Bruseth, 1987: 128). Bottomlands are considered to
be the most significant part of the entire region in terms of availability of native wild
foods in these areas (Lynott, 1977; Ferring, 1990:17). Mast crops and many plant
resources available in the bottomlands of the Eastern Cross Timbers are also present in
the bottomlands of the Blackland Prairie. A potential for a large mast crop also exists in
the upland oak woodlands of the Eastern Cross Timbers (Prikryl, 1990: 12). In contrast
to the uplands of the Eastern Cross Timbers, the Blackland Prairie uplands are
grasslands. Upland areas typically provide grains yet “seeds of the dominant grasses of
the prairie uplands are small and were probably not valuable food sources for aboriginal
people” (Prikryl, 1990: 12). However, “it is suggested that some plants in the prairie
uplands, such as forbs, are usable foods” (Prikryl, 1990: 13). Some investigators
suggest, “while human consumption of upland acorns in the Eastern Cross Timbers was
important, these mast resources were also clearly important in sustaining fauna hunted
by prehistoric people” (Prikryl, 1990: 12).
Many of these resources are available in more than one season with summer and
fall having the highest number of species for potential use. The “floral resources that
59
are most abundant between July and September are mustang grapes, wild plums,
acorns and pecans” (Prikryl, 1990: 12). Pecans are available in the fall and throughout
winter. Acorns become available in September and are plentiful through February
(Prikryl, 1990). The most abundant plant resource are these mast (nut) crops, that
would have made this region very attractive to prehistoric people, deer, squirrels and
other animals. As such, it is expected that north-central Texas offered an abundance of
resources in the fall and winter. This suggests that the region was seasonally occupied
and probably accommodated tribal aggregation.
In prehistoric times five main groups of plants were collected and eaten by various
Native American tribes: fruits, roots, seeds, greens and nuts (Steward, 1938; Ferring et
al., 1978; McCormick et al., 1975). Examples of edible fruits include plums,
blackberries, yucca, prickly pear, mesquite, clove current, mulberries, persimmons,
hackberries and grapes. Roots utilized include sunflower, milkweed, cattail, bulrush,
thistle, wild onion, potato varieties and wild buckwheat. Pecan, black walnut with burr,
blackjack, post, and black oak nuts were also eaten. Seeds are often found in
association with the grasslands, Cultures in the region were aware of the edibility and
nutritional value of these species and probably enjoyed access to these resources.
Vegetation has a timeless relationship with geology, soils, and climate that helps
form a picture of the past natural landscape. Available vegetation resources at certain
intervals effected humans, animals, and the interaction therein. In order to learn more
about human behavior in relation to the landscape inferences are made based on
geology, soils, vegetation, faunal remains and artifacts found within site-catchment
60
areas. Vegetation is nourished by the geology, soil and climatic conditions in each
physiographic region. This same vegetation transforms this energy that it has derived
from various combinations of the ecosystem into the characteristic biomass that
sustained human and faunal prehistoric populations for thousands of years.
Faunal Resources in North-Central Texas Faunal presence is directly associated with the botanical resources described in the
last section. Fauna is culturally significant in that it served as a prehistoric food source
and is an indicator of climate, seasonality and landscape. Fauna, adapted to the
Eastern and Western Cross Timbers, Grand Prairie and Blackland Prairie, provided a
variety of food resources for prehistoric populations. Previous archaeological
investigations are the main source of information for gathering data about animal
resources that would have been available during these various time periods. Faunal
data is limited to certain locations where excavations were conducted. As such
inferences are also extended to the fauna that would be drawn to a given habitat.
Verrett and Bousman (1973) view the Eastern Cross Timbers as an economically
valuable location for base camps. A variety of hunting and gathering opportunities are
provided in this area,
Small groups of deer, buffalo, and other game animals could be hunted in the small grassland openings within the Cross Timbers. This forested section would have been ideal for trapping fur bearing animals such as fox, beaver and for hunting deer. Fish, mussels and riverine fauna could be exploited along the small Rivers and creeks typical of the area (McCormick, 1975: 16).
61
Major potential faunal resources include bison, white-tailed deer, antelope, turkey,
raccoon, squirrel and rabbit. Other mammals associated with the Eastern Cross
Timbers are cottontail, jackrabbit, armadillo, opossum, beaver, skunk and black bear.
Seasonally available resources include migratory birds, especially waterfowl which
feed along the aquatic margins, spawning fishes, amphibians and reptiles, which are
more active in warmer seasons (Ferring and Yates, 1998). According to sources cited
by Prikryl (1990:11) “320 species of birds and 43 species of migratory waterfowl have
been identified in the Cross Timbers and Prairies.” Most large game species tend to
aggregate in greater numbers within the bottomland forest zone. This would have
been especially true “during fall and winter when acorns ripen and fall and when some
late fruits like persimmons ripen” (Ferring and Yates, 1998: 12).
Lynott (1977:30) believes that it was not feasible for prehistoric people to exploit
the acorn crops of the Eastern Cross Timbers uplands because, combined with their
acidic content, the acorns are smaller in size than those of the bottomlands. However,
he notes that the acorn upland mast crops were important in sustaining many species
of animals that were hunted by prehistoric people (Prikryl, 1990: 12). Flora of the
uplands would have been important in sustaining the herds of herbivores, principally
bison and antelope, hunted by prehistoric people (Prikryl, 1990: 13). Information
provided by Goodrum et al. (1971: 527-529) indicates that the acorn yields are heavily
used as food by deer, raccoon, squirrel, wild turkey and quail. Squirrels and wild
turkeys feed on acorns, and these nuts may constitute about 75% of the diet of
squirrels (Prikryl, 1990: 12). Goodrum et al. (1971) propose that acorns make up as
62
much as 50% of the diet of white tailed deer, a species that was frequently hunted by
prehistoric people (Prikryl, 1990: 12). Evidence suggests that white-tailed deer
provided the largest percentage of meat consumed by prehistoric people in the region.
Deer would have been common in the bottomland forests of both the Blackland Prairie
and the Eastern Cross Timbers, and in the uplands of the Eastern Cross Timbers. Other
species that were drawn to the mast crop also would have occurred in these three
settings (Prikryl, 1990: 13). Although fewer useful species would have been available in
the prairie uplands, bison and antelope were present.
Presumably the most important large animals of the undisturbed prairies were
bison, antelope, bear, and perhaps mostly along the wooded fringes, white-tailed deer.
Bison are reported to have been present in north-central Texas during Late Prehistoric
times (Prikryl, 1990: 13). Historic accounts indicate that bison were also present in
great numbers on the Blackland Prairie in the Brazos River area during the eighteenth
century (Newcombe, 1986). On the basis of faunal data from archaeological sites,
Dillehay (1974: 180) suggested that there were “two periods in the past--6000/5000 BC
to 2500 BC and AD 500- AD 1200/1300--when bison were essentially absent from the
southern plains and the vicinity.” In relation to this he has proposed that bison were
present on the Southern Plains before 6000 BC and between 2500 BC and AD 500.
Lynott (1979), on the other hand, believes that there is a lack of evidence to indicate
bison presence in north-central Texas during these. His review of the Holocene faunal
evidence pertaining to the sites in north-central Texas and has concluded that bison
were never very common in the tall grass prairies, but after AD 1200 their numbers did
63
increase to a certain extent. The faunal record is mainly one of terrestrial animals.
While isolated finds are common, assemblages from well-defined depositional contexts
with established taphonomies are uncommon. Faunal remains that are most commonly
reported from aboriginal sites in north-central Texas consist of deer, cottontail, turtle,
and mussels. Yet it should be noted that many sites investigated in the region have
acidic sandy loams that do not preserve bone well. Furthermore, limited attention has
dedicated to collection and identification of animal remains until recently. As such,
“animals may have made up a more significant part of the aboriginal diet than evidence
indicates” (Prikryl, 1990: 13).
Excavations conducted along the Elm Fork have provided a wealth of faunal data
where change through time can be observed (Ferring and Yates, 1997). In order to
evaluate fauna that was typically utilized during prehistoric times excavated sites from
the northern and southern portion of the Elm Fork are evaluated (figure 6). Fauna
recovered includes a variety of aquatic species such as fish, frog, turtle and beaver.
Turtle is by far the most common from this group. Species classified in previous studies
as occupying grasslands/woodlands, grasses/wooded edges, woodlands/wooded edges
and wooded edges are grouped into ecotonal settings. These fauna include some
varieties of turtle, turkey, cottontail, birds, skunk, white-tailed deer and fox. White
tailed deer and cottontail are the most numerous represented in the archaeological
record. Grassland species include lizards, squirrel, rodent, gopher, rabbit, pronghorn
and bison. Pronghorn are the most numerous from this group. Bottomland and
woodland species such as opossum, gray squirrel, snake, swamp rabbit, spiny lizard,
64
armadillo, red fox, and raccoon are grouped into the same habitat category. Wolf,
coyote, dog, snake, and unidentifiable small, medium and large mammals are grouped
into the “various” category. Deer, turtle, mussels, rabbit and turkey are the most
common fauna reported. Mussels, often found at archaeological sites in the region,
drew populations to local stream beds. However, they have “low caloric yield and have
been viewed mainly as a dietary supplement” (Prikryl, 1990: 13). Aquatic and ecotonal
settings appear to be the most heavily utilized locales in relation to food procurement
situated primarily along the Elm Fork of the Trinity. However, one site in particular in
the northernmost portion of the study area indicates a notable utilization of bottomland
fauna and grasslands rather than ecotonal aquatic zones. Unfortunately, no equivalent
data sets are available for the East or the West Forks. Only one site contains definitive
bison remains in the form of “bison scapulas” (Lynott, 1975:126). According to Lynott,
data support that subsistence strategies along the East Fork of the Trinity were based
primarily on plants and animals from bottomland-riverine zones. The variety of
resources supported by the bottomland environment, primarily include “deer, raccoon,
opossum, rabbit, squirrel, and wild turkey” (Ferring and Yates, 1998: 12). Although
currently the data are lacking there was a wide variety of faunal resources available to
prehistoric populations.
Faunal evidence recovered from archaeological sites ranges from the Middle Archaic
to the second phase of the Late Prehistoric. A map depicts the distribution of these
sites where fauna was excavated (figure 6). A graph (figure 6a) provides a summary of
collective data from sample sites regardless of time period. The purpose of viewing the
65
data in this manner is to evaluate the range of potential wildlife in each area all other
things being equal. When change through time is considered the fluctuations can be
observed to determine what types of fauna were available during specific time periods.
A series of graphs (figure 6b-figure 6c) illustrate excavation sequences. This information
is evaluated to look for correlations with hypothesized climatic conditions and to explain
settlement patterns in relation to the habitats that provided resources through time.
Data available for the Middle Archaic is limited to 41DN102 situated along the northern
portion of the Elm Fork. General trends fluctuate through time and suggest the
utilization of resources from a variety of environments. In comparison with other time
periods bottomland resources are the most dominant during the Middle Archaic. In the
earlier and later levels grassland and aquatic resources are abundant. Ecotonal
resources also follow the same trend. The middle levels at this site contain bottomland
and grassland resources in order of abundance. This pattern supports the drier climate
that has been hypothesized during this time period. Comparatively Middle Archaic
populations utilized a wider variety of resources than any other time period.
Sites that contain evidence of Late Archaic occupation include 41DN26, 41DN346,
41DN381, 41CO144, 41CO141 and 41CO150. The geographical distribution of these
sites provides a fairly balanced coverage of the Elm Fork. In all but one of these
locations aquatic resources are fairly abundant yet ecotonal resources, although
fluctuating, are continuously dominant. Site 41DN381 is the exception where a wider
range of resources are being utilized as indicated by the abundance of the “various”
category. This evidence supports that the climate was relatively wetter than the Middle
66
Figure 6. Archaeological faunal distribution from excavated sites in study area.
0% 20% 40% 60% 80% 100%
41DN197
41WS38
41COL36
41DN346
41DN102
41CO144
41DN103
41DN99
41CO141
41CO150
41DN372
AquaticGrasslandsBottomlandsEcotoneVarious
Figure 6a. Fauna recovered from archaeological sites in study area (data summarized from Lynott, 1975; Ferring, 1997, 1998; UNT 1990-2001).
67
DN27DN26 DN99
0
50
100
150
200
1 3 5 7 9 110
50100150200250
1 3 5 7 9 11
DN372
0500
1000150020002500
1 2 3 4 5 6 7 8 9
DN102
05
10152025
1 3 5 7 9 11 13
DN346
0
10
20
30
40
1 3 5 7 9
0
50
100
150
1 3 5 7 9 11 13528
DN103
0
50
100
150
1 2 3 4 5 6
CO144A
0
50
100
150
1 3 5 7 9 11 13
CO144B
0
5
10
15
1 2 3 4 5 6 7
CO150A
0
20
40
60
80
1 3 5 7 9
CO150B
0
500
1000
1500
2000
1 2 3 4 5 6
CO150C
0100200300400500600
1 2 3 4 5
1268
1680 19852750 2910
2284
LATE ARCHAIC LATE ARCHAIC
LATE ARCHAIC
CO141
0
10
20
30
40
50
1 2 3 4 5 6 7
CO141B
0
100
200
300
400
1 2 3 4 5 61282759
LATE ARCHAICLATE PREHISTORIC II
LATE PRE ILATE PRE I I
DN381
0
50
100
150
1 4 7 10 13 16 19
LATE PRE I I LA LATE PREHISTORIC II
LATE PRE I / II LA
M IDDLE ARCHAIC
610
LATE ARCHAIC
LATE ARCHAIC
LATE PREHISTORIC II LATE PRE I I LA
LATE PREHISTORIC II
Figure 6b. Archaeological fauna and change through time (Aquatic=Blue; Grassland=Green; Woodland=Brown; Bottomland=Dashed Brown).
68
CO144B
0
10
20
30
40
50
1 2 3 4 5 6 7
DN27
050
100150200250
1 3 5 7 9 11
DN26
050
100150200250
1 3 5 7 9 11
DN372
0200400600800
1000
1 3 5 7 9
DN102
0
10
20
30
1 3 5 7 9 11 13 15
DN346
05
10152025
1 3 5 7 9 11
DN99
0
50
100
150
1 3 5 7 9 11 13528
DN103
0
20
40
60
1 2 3 4 5 6
CO144A
020406080
100
1 3 5 7 9 11 13
CO150A
010203040506070
1 3 5 7 9 11 13
CO150B
0
200
400
600
800
1 2 3 4 5 6
CO150C
0
100
200
300
400
500
1 2 3 4 5
1268
1680 19852750 2910
2284
LATE ARCHAIC LATE ARCHAIC
LATE ARCHAIC
CO141
0
20
40
60
80
100
1 2 3 4 5 6 7
CO141B
0100200300400500600
1 2 3 4 5 6 71282759
LATE ARCHAICLATE PREHISTORIC II
LATE PRE ILATE PRE I I
DN381
0
50
100
150
1 4 7 10 13 16 19
LATE PRE I I LA LATE PREHISTORIC II
LATE PRE I / II LA
M IDDLE ARCHAIC
610
LATE ARCHAIC
LATE ARCHAIC
LATE PREHISTORIC II LATE PRE I I LA
LATE PREHISTORIC II
Figure 6c. Archaeological fauna and change through time
(Ecotone=Red; Various=Green).
69
Archaic in the sense that more water resources would have been available and it would
have been possible for terrestrial wildlife that was associated with wetlands to be more
mobile. Overall, there appears to be a trend that reflects fluctuating climatic conditions
due to a dramatic increase of aquatic resources followed by a decrease that is equal in
intensity. The fact that ecotonal resources are the most abundant indicates a
propensity to utilize the resources from both the grassland and woodland environments.
Only two sites contain fauna that is associated with the first phase of the Late
Prehistoric (41DN99 and 41DN381). Due to the “fuzzy” boundaries this discussion will
combine the two phases of the Late Prehistoric with more emphasis on the second
phase. Late Prehistoric II sites include 41DN26, 41DN27, 41DN99, 41DN103, 41DN346,
41DN372, 41DN381 and 41CO141. During the second phase of the Late Prehistoric
there is an interesting trend that provides a contrast to the other time periods. There
are two situations taking place. One is an inversion from the previous patterns
indicated by a dramatic shift from resources located in the ecotone to resources from
various environments. The “various” category indicates a wider range of food resources
and higher degree of mobility. In the northernmost part of the area at 41CO141
ecotonal resources are dominant throughout followed closely by aquatic resources.
Both faunal groups follow the same bell curve trend and there is a slight presence of
grassland fauna as well. To the south 41DN99, 41DN103 and 41DN346 enjoy a healthy
presence of aquatic fauna that increases and decreases through time. Sites 41DN99
and 41DN103 both maintain a fairly substantial abundance of ecotonal fauna. Site
41DN346, a little further south, follows an opposite trend and shifts to utilization of
70
fauna from the “various” category. Approximately 20 km southwest, a cluster of sites
(41DN26, 41DN27, 41DN372 and 41DN381) contain evidence similar to the other sites
assigned to this time period but with a notable abundance in woodland fauna. A
decrease in ecotonal resources and increase of various habitat echoes the pattern at
site 41DN346. During the Late Prehistoric, population, territorialism, seasonal rhythms
and fairly stable climatic conditions factor into a wide range of subsistence strategies.
Climate of North-Central Texas
North-central Texas is located in a dramatic transition between regional climates.
The striking change in vegetation from “the East Texas deciduous forests on the
eastern margin of north-central Texas to the grasslands of the plains to the west is a
vivid reflection of this climatic transition” (Diggs et al., 1999: 28). The climate of this
region has a “seasonal, subhumid precipitation regime and a strongly seasonal, thermic
temperature regime” (Ferring and Yates, 2001: 16). For the most part, “summers are
hot and winters are mild except for brief periods of cold temperatures associated with
Arctic fronts” (Ferring and Yates, 2001: 16). These fronts frequently are “accompanied
by rain, and less frequently by snow and/or ice storms” (Ferring, 2001: 16). These
general trends are thought to be fairly consistent through time.
In terms of precipitation, there is a steep East-West gradient across north-central
Texas (figure 7). Mean annual precipitation ranges from about 46 inches in the
northeastern corner of the area to about 24 inches in the Western-most portion of the
Western Cross Timbers. In general mean annual precipitation decreases about 1 inch
for each 15 miles across Texas from East to West. This region is “prone to drought
71
Figure 7. Precipitation patterns in Texas and north-central Texas.
72
FicePaD
West Fork Climate Data (1951-1980)
010203040
J F M A M J J A S O N D
Month
Tem
pera
ture
(F
)
0
2
4
6
Prec
ipita
tion
(In)
Temp Precip
Elm Fork Climate Data (1939-1969)
010
2030
40
J F M A M J J A S O N D
Month
Tem
pera
ture
(F
)
0
2
4
6
Prec
ipita
tion
(In)
Temp Precip
East Fork Climate Data (1934-1963)
0
10
20
30
40
J F M A M J J A S O N D
Month
Tem
pera
ture
(F
)
0
2
4
6Pr
ecip
itatio
n (In
)
Temp Precip
gure 8. Average monthly temperature and precipitation trends in north-ntral Texas (compiled for West Fork (Ressell, 1983); Elm Fork (Ford and uls, 1978); and East Fork (Hanson and Wheeler, 1969) from Wise,
enton, and Collin County Soil Surveys respectively).
73
while at other times it receives too much rain in a short time” (Diggs et al., 1999: 29).
When the climate patterns occurring in the area around the West Fork, Elm Fork and
East Fork are compared (figure 8) the differences are minimal. The most notable
difference is the increased temperature and decreased precipitation, in the East Fork in
December. The opposite situation is true for the Elm Fork whereas both temperature
and precipitation are relatively low in the West Fork portion of the study area.
Combined Climate Data
0
5
10
15
20
25
30
35
J F M A M J J A S O N D
Month
Tem
pera
ture
(F)
1
2
3
4
5
6
7
8
9
10
Pre
cipi
tatio
n (In
)
Wise Temp Denton Temp Collin TempWise Precip Denton Precip Collin Precip
Figure 9. Comparisons of Average Monthly Temperature and Precipitation Trends in north-central Texas (compiled for West Fork (Ressell, 1983); Elm Fork (Ford and Pauls, 1978); and East Fork (Hanson and Wheeler, 1969) from Wise, Denton, and Collin County Soil Surveys respectively).
74
In general, precipitation and temperature data (1931-1989) indicate that late spring
and early fall are the wettest months while summer temperatures are high
accompanied by a decrease in rainfall (figure 9). These historical records indicate the
average rainfall as 32 inches per year and the average temperature is 65.2 F. The
similarities in climatic data for the three counties in the study area indicate that the
climate in general was fairly uniform with minor variations on the theme.
Past Environments
The entire eco-system is interdependent and certainly the climate was a major
factor for decision making in the past. Past environments have been extrapolated from
pollen and other organic data preserved by stratified profiles. The Ferndale Bog pollen
record is the sole source of early Holocene vegetational history for the region. Given
the proximity to the study area combined with the presence of these vegetation families
it is inferred that similar patterns likely occurred in north-central Texas during these
time periods. For the current research, this has been taken into consideration and
these data are mapped to provide a view of change through time in relation to the
relative abundance of these taxonomic families.
From a long-term perspective, pollen, plant macro fossils, and other types of
evidence demonstrate that the climate of Texas has changed substantially over the past
15,000 years (Diggs et al., 1999: 30). Paleoenvironmental change is not well
documented, but it is summarized by Prikryl (1993: 192-193) and updated by Ferring
and Yates (1997: 39-45). Prior to 14,000 BP the climate of north-central Texas was
cooler and moister than at present. Between 14,000 and 11,000 BP, the climate
75
became drier. It is suggested that the presence of high grass pollen and low arboreal
pollen between 7550-3050 BP shows a drying return of arboreal pollen after 3050 BP.
The latter change is similar to today’s environment. High grass pollen also occurs at
approximately 1550 BP and from 3550 to 3650 BP, and this also indicates drier periods.
The presence of paleosols between 2000 BP and 1000 BP suggests a decrease in
moisture with a return to wetter conditions after 1000 BP (Skinner and Ferring, 1999).
Prehistoric climatic data is in great need of improvement. Palynological data
supplemented by data from insects has the potential to add a tremendous amount of
breadth to the current research methodology and resulting settlement model.
This chapter has reviewed the geology, topography, hydrologic networks, soils,
physiography, vegetation, faunal communities, and general climatic trends that have
given the environment in north-central Texas its character over the past 10,000 years.
The environmental foundation is half of the entire ecological system. The other half of
this story is completed by the human populations that inhabited the landscape. A
chronological overview of these cultures provides a means to observe change through
time and establish a set of expectations regarding the range of behavior that has been
supported by previous archaeological evidence.
76
CHAPTER 4
CULTURAL CHRONOLOGY OF NORTH CENTRAL TEXAS
This chapter defines the characteristics of prehistoric populations, in north-central
Texas, as they change through time. Previous research has established specific
characteristics that define cultural subdivisions. Artifacts are a primary source of
differentiation between cultural groups and are utilized as an indication of time period
for surface sites that lack radiocarbon dates. Therefore, determining the range of
“diagnostic” projectile point types and character of the tool kit, for each time period, is
pertinent to the current research. Artifacts are a significant source of information
related to prehistoric activities. Subsistence evidence facilitates comparisons between
time periods and serves as an explanatory mechanism for site location or variable
density of occupation. Settlement pattern behavior, as discussed by previous
investigators, provides a means for comparison. These data collectively establish a
cultural framework to relate to the results of the SCA conducted on archaeological sites
dating from the Early Archaic to the Late Prehistoric.
North Texas prehistory spans across approximately 12,000 years. Geographic
distributions of prehistoric cultures reveal that ancient populations inhabited virtually all
of Texas with the exception of a few counties where evidence has yet to be reported.
Based on data provided by a variety of sources (THC, 1985; Skinner, 1985; Ferring,
1997, 1998; TARL 2002), a series of maps have been built to illustrate the change in
prehistoric Texas population distribution through time, with respect to physiographic
77
boundaries, the north-central Texas region and the research study area. The first of
these maps illustrates all prehistoric populations of record (figure 10).
Figure 10. Distribution of prehistoric sites in Texas and north-central Texas (data compiled from Nunley, 1973; Lynott, 1977; Biesaart et al., 1985; Prikryl, 1990; Ferring et al., 1997, 1998, 2001; TARL, 2002).
78
Chronologies that have been developed for north-central Texas vary according to the
investigator and reflect the lack of clarity or consensus in relation to cultural
subdivisions. These chronologies (table 1) have been developed and correlated with
projectile point typologies (figure 11). There is disagreement as to the correlation
between the precise time period, point typology and cultural subdivisions. However, in
general, there seems to be a certain level of agreement as to the general parameters of
the various time periods. This is a geographically related issue that depends on the
amount of work done in a certain region and is relative to few sites that have been
radiocarbon dated. The overlap between time period designations and the various
labels given to each time period by different investigators are important to note when
comparing archaeological reports and corresponding regional models.
Based on stratified excavations “point types” have been established as “index
fossils” to define each time period. Notable chronological sequences of point types that
have been developed in Texas (Prewitt, 1981; Prikryl, 1990; Story, 1990) have served
as the main reference for archaeological investigations. An overview of the currently
accepted point type chronology for this research is illustrated pictorially (figure 11).
These point types were recovered from archaeological sites, included in the current
research, and are associated with time periods based on radiocarbon-dated
investigations (figure 12). Archaeological data available for sites in north-central Texas
is contained in a variety of contract reports, academic publications, journal articles and
site reports filed with Texas Archaeological Research Laboratory (TARL).
79
Chronology
Archaic Neo-American McCormick et al. 1975
9000 BP-1500 BP 1500 BP- 450BP
Chronology Archaic
8000 BP-1200 BP Neo-American 1200 BP- 350BP
Skinner et al. 1985
Carrollton Focus
8000 BP- 4500 BP
Elam Focus 4500 BP-1200 BP
Wylie Focus 1100 BP-400 BP
Henrietta Focus 700 BP-250 BP
Chronology Early
Archaic Middle Archaic
Late Archaic
Johnson and Holliday, 1986:7
8000 BP- 6000 BP
600BP- 4000 BP
4000BP-2000 BP
Chronology Early
Archaic Middle Archaic
Late Archaic
Late Prehistoric I
Late Prehistoric II
Ferring et al. 1990
8500 BP- 6000 BP
6000 BP- 3500 BP
3500 BP- 1300BP
1300 BP- 800 BP
800 BP- 350 BP
Chronology Early
Archaic Middle Archaic
Late Archaic
Late Prehistoric I
Late Prehistoric II
Lebo and Brown 1990; Prikryl, 1990
8500 BP- 6000 BP
6000 BP- 3500 BP
3500 BP- 1250BP
1250 BP- 750 BP 750 BP- 250 BP
Chronology Early
Archaic Middle Archaic Late
Archaic
Late Prehistoric
Ferring et al. 1997
8500 BP-6000 BP
6000 BP- 3500 BP
3500 BP- 1250BP
1250 BP-350BP
Table 1. Various chronologies developed for cultural subdivisions.
80
Fpfr
igure 11. Overview of Early Archaic through Late Prehistoric I generalized projectile oint chronology in north-central Texas that is utilized for current research (compiled om Turner and Hester, 1993; illustrations used with permission).
81
Time Period
Early Archaic
Middle Archaic
Late Archaic
Late Prehistoric I
Late Prehistoric II
Morrill
Wells
Marshall
Tortugas
Palmillas
Carrollton
Castroville
Frio
Dallas
Ensor
Elam
Ellis
Trinity
Gary
Edgewood
Godley
Yarborough
Alba
Kent
Scallorn
Fresno
Bonham
Perdiz
Steiner
Washita
Catahoula
Maud
Toyah
Figure 12 Chronological overview of projectile point types from north-central Texas (data compiled from Suhm and Jelks, 1962; Turner and Hester, 1993).
82
Paleoindian (11,500-8,500 BP)
Archaeologists generally agree that the first people that migrated into Texas arrived
around 12,000 years ago around the end of the Pleistocene. These early inhabitants
left behind distinctive artifacts that reveal sophisticated tool manufacturing techniques.
Although Paleoindian artifacts have a widespread geographical distribution in Texas
most evidence is found on the surface (figure 13). In north-central Texas the only well
documented “in situ” excavation and subsequent publication, regarding a Paleoindian
culture in north-central Texas, is the Aubrey Clovis site (Ferring, 2001). Otherwise,
Paleoindian sites in the north-central Texas region are predominantly surface exposures
in mixed association with later occupations. The current research does not include
analysis of Paleoindians but it is necessary to review the characteristic artifacts, lifeways
and settlement patterns associated with these populations because they essentially set
the stage for the Archaic cultures.
Paleoindian cultures are defined by specific distinguishable characteristics that
differentiate them from all of the other time periods. The single most distinctive trait of
the Paleoindian cultures is the presence of unique and finely crafted projectile points.
Clovis, Folsom, Midland, Plainview, Firstview, Eden, Angostura, Scottsbluff, Dalton, San
Patrice and Golondria are projectile points (figure 14) associated with the Paleoindian
period. Of the forty-two projectile points, documented by Prikryl (1990): Clovis, Dalton,
Golondria, Midland, Plainview and Scottsbluff provide evidence of a Paleoindian
occupation in north-central Texas. The Paleoindian period has been defined by three
horizons based on projectile point typology: Clovis, Folsom, and Plano respectively.
83
Figure 13. Distribution of Paleoindian sites in Texas and north-central Texas (data compiled from Nunley, 1973; Lynott, 1977; Biesaart et al., 1985; Prikryl, 1990; Ferring et al., 1997, 1998, 2001; TARL, 2002).
84
The Clovis horizon is primarily represented by finely crafted points with basal flutes and
laurel-leaf shaped bifaces. Tool kits include a limited variety of tools consisting of
blades, scrapers, and bone or ivory cylindrical shafts (Lynott, 1981). It is common for
this time period that imported raw materials, rather than locally available quartzites and
cherts, were utilized in the manufacture of projectiles and tools (Ferring, 2001).
Utilization of non-local raw materials indicates increased mobility for long-distance
procurement by direct acquisition or evidence of cultural interaction.
Tool assemblages associated with the Folsom horizon include fluted and non-fluted
projectile point forms and various stages of lithic reduction that reveal a predominant
bifacial technology. Scrapers, choppers, perforators, hammerstones, abraders and bone
needles are also characteristic tools (Story, 1990: 189). Shifts in technology are
indicated by an increased frequency of unfluted points, such as, the Plainview, Midland,
and Eden. As with Clovis horizon artifacts, high-quality lithic raw materials were used
for Folsom projectile points, implying long-distance travel or trade.
Plainview, Eden, Meserve, Angostura, Scottsbluff, Dalton and San Patrice point types
are commonly associated with the Plano horizon. In correlation with other regions these
points indicate a date around 10,000 BP. Dalton and San Patrice points are associated
with the Paleoindian period. However, lifeways suggest affinities with the Early Archaic
period. Therefore these particular projectile points are considered to be part of a
transition between Paleoindian and Early Archaic technologies (Lynott, 1981:104;
Prikryl,1987:157). Given that many of these point types were discovered in the study
area, the Plano horizon is particularly important to the current research.
85
Figure 14. Paleoindian Projectile Point Chronology for north-central Texas (compiled from Turner and Hester, 1993; illustrations used with permission).
86
Surveys conducted along the Elm Fork of the Trinity by Prikryl (1990) mainly recovered
Plainview and Dalton varieties. Dalton tool kits contain a variety of scrapers, retouched
pieces, unifacial tools, blades, blade-like flakes, bipolar cores, wedges and specialized
tools such as the uniquely fashioned bifacially chipped stone adze that was probably
employed for woodworking (Story, 1990:190-202). Unlike the Clovis and Folsom the
Plano horizon encompasses a more diverse set of cultural components as characterized
by a proliferation of projectile point types and tools.
Paleoindian cultures hunted extensively in a fairly generalized manner from very
small species to megafauna. Many Paleoindian sites contain remains of large mammals
such as: mammoth, mastodon, bison, caribou, deer, camel, and horse. Faunal analyses
at the Aubrey Clovis site indicate broad exploitation of large, medium, and small game,
including bison, deer, rabbits, squirrels, fish, and turtle (Ferring and Yates, 1997). The
range of variation presented by the faunal data indicates very flexible adaptive
strategies (Ferring and Yates, 1997). Many of the large mammals found in association
with Clovis sites became extinct requiring new adaptations. Folsom populations
adapted adjusting their lifeways to respond to the challenges presented by these
changes that are often attributed to the natural environment. This adjustment included
a shift to hunting herds of Bison, refined tools and new techniques. The Plano horizon
is believed to be the "final big game hunting occupation of the southern plains" (Lynott,
1981: 101). In north-central Texas, evidence is lacking for a subsistence strategy
based heavily on big game hunting. A generalized hunting and gathering lifeway is
considered more likely for Paleoindian populations that inhabited this region.
87
Vegetation during the Clovis horizon in this region was transforming due to
decreased rainfall and warmer temperatures toward the end of the Pleistocene. Clovis
point distribution suggests that all major environmental zones were exploited at one
time or another (Story, 1990: 178). Around 9,200 BP the Folsom horizon developed in
association to climatic changes and a complete shift to the hunting of bison. Shortgrass
prairie ecosystems developed during this time period and were preferred by bison
(Lynott, 1981: 101). During the Plano horizon evidence suggests that populations
associated with Dalton points utilized woodland edge and prairie environments whereas
San Patrice populations are associated with a more restricted range of environments.
In general, the Paleoindian period is characterized by a series of flexible adaptations
that correspond to the changing environmental conditions of the Late Pleistocene.
Paleoindian Settlement Patterns (11,500-8,500BP)
The settlement pattern for Clovis horizon populations is focused on upland settings
and along small tributary streams (Story, 1990:182). In general, Clovis populations in
north-central Texas were highly mobile and traveled considerable distances in small
bands (Lynott, 1981: 101; Johnson 1989), indicating an overall low population density.
Folsom presence in the region appears to have been very limited (Story, 1990: 189).
Towards the end of the Paleoindian era, it has been speculated that a slight reduction
of group mobility occurred in north-central Texas, possibly due to an increase in the
tallgrass prairies coupled with an increase in population density (Lynott, 1981:101-103).
However, repeated occupation of late Paleoindian sites by Early Archaic cultures
obscures whether or not the increased population densities can be attributed to Plano
88
horizon cultures. Nevertheless, the Plano horizon is mainly associated with Dalton points
which are fairly numerous in north-central Texas suggesting more intensive use of the
area in the Plano horizon than during the Clovis, Folsom or Early Archaic. Lithic raw
materials indicate that the settlement patterns of these groups focused on primary and
sometimes far-removed sources such as Alibates, Edwards Plateau, and the Ouachita
Mountains supporting a significant level of mobility (Prikryl, 1990; Banks, 1990). Data
from these sites combined with the lack of permanent structures indicate the existence
of small, highly mobile groups utilizing scattered campsites. Overall, higher population
densities, a reduction of mobility, and an emergence of territories are indicated towards
the close of the Paleoindian era (Story, 1990: 196). Although notably different in
character, hunting and gathering lifestyles endured throughout the Archaic.
Archaic (8,500-1,250 BP)
The Archaic time period represents a series of adaptations that were very similar to
the Paleoindians (McGregor, 1987). The geographical distribution of Archaic cultures
provides insight into the overall population density in Texas as it relates to the north-
central Texas region (figure 15). Patterns indicate a fairly widespread distribution and
increased population as suggested by the increase of sites reported across the state.
The Archaic period has been defined as a time period designation for hunter and
gatherer populations who practiced significantly different life-ways than their
predecessors in relation to mobility and subsistence (Story, 1990:211). As previously
mentioned, the varying chronologies produced by investigators for North Texas reflect
the lack of clarity as to cultural subdivisions of the Archaic. However, the number of
89
Archaic period subdivisions primarily reflects differences in the intensity and
geographical concentration of research conducted in each region (McGregor, 1987).
Figure 15. Distribution of Archaic sites in Texas and north-central Texas (data compiled from Nunley, 1973; Lynott, 1977; Biesaart et al., 1985; Prikryl, 1990; Ferring et al., 1997, 1998, 2001; TARL, 2002).
90
The more generalized chronologies reflect a lack of well-stratified sites and dated
components. One of the main difficulties with Archaic age sites is the tendency for the
mixing of cultural layers and frequency of surface finds that may be related to site
formation processes or re-use of lithic materials in areas of low sedimentation.
However, a handful of stratified sites have contributed to the development of
chronologies based on projectile point types found in radiocarbon-dated strata. The
current research depends on a chronology that has been most commonly accepted and,
in part, based on radiocarbon dating in north-central Texas. This provides a date for
the Archaic between 8,500 BP and 1,250 BP (Ferring and Yates, 1998).
Varieties of point styles (figure 16) that emerge during the Archaic provide a view of
the tremendous diversity that bloomed and brought the Paleoindian period to an end.
Rather than a small assortment of finely crafted artifacts, Archaic cultures adapted by
developing an extensive assortment of stone tools. In addition to a variety of
projectiles, scrapers, knives, gravers, burins, drills, gouges and axes were added to the
toolkit (Lynott, 1986; Story, 1990; Ferring and Yates, 1998). During the Archaic a
variety of small, stemmed dart points are first evident and polished stone tools appear
for the first time along with grinding implements, mortars, manos, pestles, and a variety
of bone tools (Ferring and Yates, 1998; Story, 1990). Hearths, middens and wells are
significant features that are heavily utilized during the Archaic (Johnson and Holliday,
1986; Story, 1990). Archaeological remains from this time period reflect clear changes
in cultural adaptations and increasing diversity in artifact assemblages suggest an
increased dependence on plants. The arrival of this diversity is probably related to
91
adaptations in response to population growth indicated by a substantial increase in
sites, increased knowledge regarding effective plant utilization and fluctuating climates.
Figure 16. Overview of Archaic projectile points in north-central Texas (compiled from Turner and Hester, 1993; illustrations used with permission).
92
Early Archaic (8,500-6,000 BP)
The geographical distribution of Early Archaic cultures (figure 17) indicates a
population increase as compared to the Paleoindian. In north-central Texas, the Early
Archaic period is very difficult to define due to the lack of well-documented, temporally
unmixed cultural deposits and use of inconsistent artifact typologies (Story, 1990:217).
To date, few sites with discrete Early Archaic components have been investigated in this
region, with the exception of Prikryl's (1990) identification of 57 diagnostic Early Archaic
artifacts in assemblages from 22 surface sites in the Upper Trinity basin.
Compared to Paleoindian period artifact assemblages, Early Archaic (8,500 BP-6,000
BP) flaked stone tools tend to be less carefully fashioned and less often made of
extralocal raw materials (Story, 1990:213). Point types that define this time period vary
according to geographic location. Evidence of Early Archaic occupations in the Ray
Roberts Lake area consists of surface finds of the Angostura, Wells and early split
stemmed projectile point types (Prikryl, 1987: 158-161). The general evolutionary
trend in projectile point manufacture (figure 18) throughout the period is toward
shorter triangular points. Compared to the Paleoindian period, the tool kit of the Early
Archaic is more diverse. Projectile points are primarily produced from local cherts and
quartzites. Appearance of polished and ground stone items suggests changes in the
processing of food resources and most likely the intensification of plant food collection
(Story, 1981; Ferring and Yates, 1998). In general, the Early Archaic is characterized by
adaptations to the modern fluctuating climate and environment that emerged during
93
the Holocene. Cultural responses to these new ecological systems include a wider
variety of projectile points and more variable artifact assemblages.
Figure 17. Distribution of Early Archaic Sites in Texas and north-centralTexas (data compiled from Nunley, 1973; Lynott, 1977; Biesaart et al., 1985; Prikryl, 1990; Ferring et al., 1997, 1998, 2001; TARL, 2002).
94
Figure 18. Early Archaic projectile points in north-central Texas (compiled from Turner and Hester, 1993; illustrations used with permission).
95
Middle Archaic (6,000-3,500 BP)
The geographical distribution of the Middle Archaic (figure 19) indicates a fairly wide
distribution with an overall decreased population density. The Middle Archaic is defined
essentially as a continuation of Early Archaic life-ways, with a few changes in adaptive
strategies. Middle Archaic sites are rare in this region (Ferring and Yates, 1997). In
the Elm Fork Trinity Valley, projectile points of this general age are found at fewer
localities than are points of any other age (Prikryl, 1990). The only “in situ” site of this
age, discovered in the study area, is 41DN102, the Calvert site (Ferring and Yates,
1997). Deep burial, dry climates and reduced occupation potentials are probably
important factors in the small number of sites dating to this period.
The growing diversity reflected in the material remains is the defining
characteristic that distinguishes the Middle from Early Archaic time periods. Generally,
projectile point size increases from the Early Archaic to Middle Archaic. In north-central
Texas, the Middle Archaic is characterized by an increased variety of projectile points
(figure 20). Dart points (e.g. Carrollton, Elam, Morrill, Wells and Gary); basal notched
projectile points (e.g. Andice and Calf Creek); and parallel, slightly contracting based
points (e.g. Pedernales, Bulverde, Travis and Nolan) are characteristic of this time
period (Prikryl, 1987: 162; Lebo and Brown, 1990). Regardless of the paucity of sites,
enough archaeological evidence has been discovered to formulate two cultural
subgroups referred to as the Trinity Aspect and Elam focus.
The Trinity aspect, which contains the early Carrollton focus, is one of the earliest
well-defined Middle Archaic period cultural entities centered along the Trinity. The
96
Carrollton focus is defined solely on the basis of artifacts and primarily projectile points.
The vast majority of the points are medium-sized to small in length. There is also
tremendous variation in the sizes of the individual point types such as the Gary point.
Figure 19. Distribution of Middle Archaic sites in Texas and north-centralTexas (data compiled from Nunley, 1973; Lynott, 1977; Biesaart et al., 1985; Prikryl, 1990; Ferring et al., 1997, 1998, 2001; TARL, 2002).
97
Figure 20. Middle Archaic projectile points in north-central Texas (compiled from Turner and Hester, 1993; illustrations used with permission).
98
A few of the Carrollton focus projectile points are similar to Plainview and others
resemble the Gary types of the Late Archaic period. Other points associated with this
"focus" include the Carrollton and Trinity dart points (McGregor, 1988:31). The tool kit
includes spokeshaves, the "Waco sinker", the chipped stone Carrollton axe, scrapers,
gouges, a few grinding stones, ground stone, drilling tools, knives, hammerstones and
choppers (McGregor, 1988:31).
The most commonly recovered artifacts from Elam focus components are projectile
points. Possibly the most common projectile point of this period is the Ellis dart point,
followed by the Yarbrough and Elam types (McGregor, 1988:31). Similar to the
Carrollton focus, drills, scrapers, knives, hammerstones and choppers remain a
significant part of the tool kit. In comparison with the Carrollton focus, the Elam focus is
characterized by a higher percentage of lithic tools made from locally obtained quartzite
(McGregor, 1988:31). Overall, however, raw materials commonly used for Middle
Archaic tools derive from locally available lithic resources. Typology is complex and
often ambiguous requiring constant reevaluation. Nevertheless, the currently
established descriptions of Middle Archaic typological characteristics offer an adequate
overall cultural model for this time period.
Late Archaic (3,500-1,250 BP)
The geographical distribution (figure 21) of Late Archaic cultures indicates increased
population densities across the state. Significantly more Late Archaic sites have been
investigated in north-central Texas, than in previous time periods (McGregor and
Bruseth, 1987; Prikryl, 1990; Peter and McGregor, 1988; Ferring and Yates, 1997). The
99
abundance of sites is attributed to shallow burial below floodplains that has produced
in-situ sites with well-preserved living surfaces, features and biotic remains. These sites
have been found to be very common along the Trinity (Ferring and Yates, 1997:6).
Intriguing evidence for Late or Transitional Archaic use and occupation of the region is
related to large man-made depressions found at the Richland/Chambers Reservoir
(Story et al., 1990; Lynott, 1975). These depressions, referred to as "Wylie Focus pits,"
are features that may indicate group aggregation and ceremonialism (Lynott, 1975;
Bruseth et al., 1987; Story, 1990:235). Several sites excavated along the Elm Fork also
provide excellent stratigraphic records of Late Archaic cultures.
Most evidence for the presence of Late Archaic cultures (figure 22) includes
projectile points such as the Ellis, Edgewood, Elam, Gary, Dallas, and Godley varieties,
as well as, the slightly earlier Yarbrough and Trinity types recovered from
archaeological contexts (Prikryl, 1987; Brown and Lebo, 1991). Late Archaic
occupations in the Lewisville Lake area are primarily based on the surface recovery of
these point types with the exceptions of the excavations reported by Ferring and Yates
(1997). New developments in the material culture for this time period include the use of
pecked and polished stone in the production of ornaments and implements such as
spear-throwing weights and axeheads (Story, 1990:213). Prikryl’s (1990) analysis of
raw material types indicates that 39% of the Late Archaic tools were made from non-
local chert whereas Early-Middle Archaic and Late Prehistoric assemblages average
70%-80% chert raw materials. Intensive use of local materials is characteristic of the
Late Archaic period (Ferring and Yates, 1997:6).
100
Figure 21. Distribution of Late Archaic sites in Texas and north-centralTexas (data compiled from Nunley, 1973; Lynott, 1977; Biesaart et al., 1985; Prikryl, 1990; Ferring et al., 1997, 1998, 2001; TARL, 2002).
101
Figure 22. Late Archaic projectile point chronology for north-central Texas (compiled from Turner and Hester, 1993; illustrations used with permission).
102
Archaic Climatic Change (8,500-1,250 BP)
The Archaic is a time period marked by climatic change from the end of pluvial
conditions through extreme xeric conditions (Johnson and Holliday, 1986: 46). These
conditions led to the modernization of faunal and vegetational communities. Beginning
in the Early Archaic period the gradual warming trend posited toward the end of the
Paleoindian period significantly changed during the early Holocene toward the
increasingly warmer, drier conditions of the Middle Holocene (Johnson and Holliday,
1986: 46). Although the trend towards a warmer, drier environment began in the Early
Holocene, wetter climates predominated during the Early Archaic (Ferring and Yates,
1997). Grasses were dominant between 9,000 and 5,000 BP (Ferring, 1993; Prikryl,
1987:156) effecting the options and related choices these early populations made.
Opening of new prairie uplands and retreat of hardwood forests to the floodplains is
related to these climatic conditions (Lynott, 1981:103). Drier climates and decreased
biotic resources are suggested during the Middle Archaic due to the presence of high
grass pollen and low arboreal pollen between 7500 and 3000 BP (Ferring and Yates,
1998:36; Skinner and Ferring, 1999:10). Beginning 5000 BP, the north-central Texas
environment underwent an increase in effective moisture and temperature that would
have affected the variety of available plants. A return to wet conditions toward the end
of this time period led to the beginning of the Late Archaic. Around 4,500 BP evidence
suggests that a migration of oak savannah may have replaced a large portion of the
grasslands (Prikryl, 1987:162). By 4000 BP the climate began to approach present
conditions. The environment, as indicated by paleoenvironmental evidence, obtained
103
from archaeological sites, experienced a slight shift to wetter conditions during the Late
Archaic period. Additional supporting evidence of wetter conditions includes the
development of the West Fork Paleosol (Ferring, 1986: 112). With wetter
environmental conditions, it is expected that the density of vegetal resources would
have increased, resulting in the expansion of the Cross Timbers woodlands. This would
have resulted in an increase of the plants utilized for human consumption and biomass
for game animals (Prikryl, 1987:170). The Late Archaic occurs during a return to
moister and somewhat cooler conditions, perhaps similar to those of today (Johnson
and Holliday, 1986: 47). However, climatically, the warm wet conditions continued for
another 1000 years. Cultural adaptations throughout the Archaic appear to correspond
with changes in climate. As such, it appears that the Archaic cultures were most
dramatically affected by the fluctuating climatic conditions.
Archaic Subsistence Patterns (8,500-1,250 BP)
Throughout the Archaic in north-central Texas economies were based on hunting
and gathering (McGregor, 1987). The subsistence pattern of the Early Archaic period is
very similar to that of the Paleoindian period. Early Archaic cultures were also nomadic
seasonal hunters except the annual range utilized by these cultures appears to be
considerably reduced. In addition, emphasis appears to have been placed more on
gathering a wider range of plant species within this annual range (Story 1981:144;
Brown et al. 1987:5-6). Extinct bison are present into at least the beginning of the
Early Archaic. During the Middle Archaic data indicates an increased modern bison
economy and a diffuse subsistence economy that included a wide variety of available
104
resources (Brown and Lebo, 1991:8). Faunal data from sites in north-central Texas
indicate that Late Archaic populations exploited a mix of prairie, forest and riparian
species (Ferring and Yates, 1997:6). According to data generated by excavations white-
tailed deer, rabbits, turtles and mussels are the most common (Ferring and Yates,
1997). In addition, mast crop resources would have been particularly desirable to these
early populations. Although the importance of hunting appears to be generally reduced
throughout the Late Archaic subsistence evidence indicates continued hunting of bison
and antelope in the upland prairie (Prikryl, 1990; Brown and Lebo, 1991). Procurement
of riverine faunal and floral resources in the prairie, Cross Timbers and ecotones
involved intensive exploitation of local bottomland resources and seasonal activities
(Lynott, 1981:104; McCormick, 1990). In some cases subsistence economy is focused
on the oak-hickory forests found along the floodplains of major drainages (Brown and
Lebo, 1991:7). As a result of a more diverse economy, site types within the Archaic
period are more varied than in the Paleoindian period. Site types include: open camps,
bison kills, flint quarries and flint caches (Story, 1990: 211). In general, Archaic
populations were hunter-gatherers that utilized a wide variety and abundance of wild
food resources.
Early Archaic Settlement Patterns (8500-6000BP)
It is probably due to wetter conditions during the Early Archaic that sites are
primarily found on stream terraces near primary water sources rather than bottomland
settings (Skinner and Ferring, 1999:9). According to Prikryl (1990) there are fewer
bottomland sites reported on the Elm Fork than during the previous period. However,
105
Lynott (1981) suggests that there was an increased emphasis on the use of bottomland
food resources (Skinner and Ferring, 1999:9). Subsistence patterns indicate a diffuse
hunting and gathering strategy on the floodplains and in the prairie uplands (Brown and
Lebo, 1991:7; Lynott, 1981:103). Given the various climatic interpretations the range
of settings occupied during this time period could be an indication of a fluctuating
climate or seasonal occupation in relation to drier or wetter conditions. The
aggregation of Early Archaic period populations appears to have been confined to small
groups of nomadic seasonal hunters who utilized briefly occupied hunting camps and
followed seasonal rounds that focused on a variety of resident and migratory animals
along the main stem of the Trinity and its larger tributaries. In general, previous
investigations indicate an Early Archaic concentration in the floodplains, bottomlands,
uplands and along major tributaries. Site locations are situated in both the prairies and
cross timbers. However, no trends or patterns have been identified, on a regional
scale, in relation to physiographic or topographic settings, to provide a clear explanation
for site location.
Middle Archaic Settlement Patterns (6000-3500BP)
The few Middle Archaic sites that have been located are on the first terraces above
the flood plains (Skinner and Ferring, 1999:9). These sites tend to be situated along
the Elm Fork rather than the smaller tributaries (Skinner and Ferring, 1999:9). Evidence
suggests that the subsistence economy was generally focused on the Oak-Hickory
forests found along the flood plains of major drainages while other sites along the Elm
Fork are located on terraces above stream floodplains (Brown and Lebo, 1991:7; Prikryl,
106
1990). Both cases indicate a direct correlation with major water sources that would
have been highly desirable during the drier climatic conditions. As scheduling of
hunting and gathering became more efficient, the size of territories became smaller
(Brown and Lebo, 1991: 8). During certain seasons, it was possible for small social
aggregates to form without exhausting local resources. Yet it would have been
advantageous for these larger groups to disperse during times of food stress (Lynott,
1981: 104). The bison hunt requires large areas of land, as well as, a strategic system
to find herds, corral and capture them. Seasonal scheduling of activities was necessary
to synchronize with the migratory patterns and cyclical abundance of herds. However,
this pattern has not been clearly illustrated in relation to site locations as they
correspond with the necessary settings that would fulfill scheduling requirements while
sustaining prehistoric populations.
Late Archaic Settlement Patterns (3500-1250BP)
During the Late Archaic there appears to be a notable shift of site location to
floodplains, first order tributaries and minor tributary streams (Prikryl, 1990: 74;
Skinner and Ferring, 1999:9). The development of the West Fork Paleosol toward the
end of the Late Archaic reflects a wetter environment that would have influenced the
range of settings appropriate for human occupation (Prikryl, 1990: 74; Brown and Lebo,
1991:8). Subsistence strategies indicate that a wide variety of settings (e.g. uplands,
riparian corridors, bottomland hardwood areas) were utilized. An expansion of the
Eastern Cross Timbers associated with the wetter environment would have provided a
larger mast crop for consumption by humans and animals (Brown and Lebo, 1991:8).
107
Evidence suggests that subsistence in the upland prairie continues to focus on bison
with some use of riverine faunal and floral resources (Brown and Lebo, 1991:8). Along
major drainages, the subsistence economies reflect increased efficiency in the use of
local floodplain resources (Brown and Lebo, 1991:8). This efficiency was probably
related to better scheduling and exploitation technologies (Brown and Lebo, 1991:8).
Although Late Archaic sites are distributed over a larger area and appear to have been
larger or more frequently occupied there is no evidence to supports that sedentary
villages existed during this period. Settlements appear to correspond with locations
that contain access to abundant resources, with favored sites being repeatedly used
over long periods of time (Brown and Lebo, 1991:8). A few sites contain clear evidence
of repeated occupations and sites with higher artifact densities are thought to indicate
repeated use as well (Ferring and Yates, 1997:6). Evidence suggests that camps and
habitation areas were not occupied year-round yet group movements appear to be
within particular territories (Story, 1981:146). In north-central Texas, the Late Archaic
settlement patterns indicate increasing evidence for the reuse of campsites and a trend
toward seasonal aggregation of populations when large quantities of food were locally
available (Lynott, 1981: 105). Higher site densities and increased cultural complexity
accompany the end of the Late Archaic.
Late Prehistoric (1250-250BP)
The classification of sites on a statewide scale combines the two phases of the Late
Prehistoric (as defined in north-central Texas) into one time period. Therefore the
generalized geographical distribution (figure 23) has been combined to form one overall
108
view. The appearance of the bow and arrow in the archeological record is the hallmark
of the beginning of the Late Prehistoric period (Story, 1990; Ferring and Yates, 1998).
Figure 23. Distribution of Late Prehistoric sites in Texas and north-central Texas (data compiled from Nunley, 1973; Lynott, 1977; Biesaart et al., 1985; Prikryl, 1990; Ferring et al., 1997, 1998, 2001; TARL, 2002).
109
In addition, an increased dependence on plants led to the use of ceramics followed by
the development of agriculture and a relatively sedentary lifeway. Prehistoric
populations appear to have been receptive to these changes as suggested by the
technological changes indicated by the archaeological record.
Although, Prikryl (1990) and others divide the late Prehistoric into two phases, Late
Prehistoric I (1,250BP-750 BP) and Late Prehistoric II (750BP-350BP), based on
projectile point types, “the number of sites with discrete assemblages is very small, and
the great majority of the Late Prehistoric sites that he identified have points assigned to
both phases” (Ferring and Yates, 1998). In the later parts of this period, Plains Village
traits are found to be more common in assemblages from the Elm Fork Trinity (Ferring,
1986) while Caddoan traits are more common in sites from the East Fork Trinity
(Lynott, 1975; 1981). As a result, in Late Prehistoric times both temporal and cultural
factors contribute to archaeological complexity. Excavations of several Late Prehistoric
sites at Lake Lewisville and Ray Roberts Lake and other local reservoirs (Skinner and
Baird, 1985; Peter and McGregor, 1988; Lynott, 1975) provide examples of in-situ
archaeological sites (Ferring and Yates, 1997:7).
Scallorn, Rockwall, Catahoula, Bonham and Alba arrow point types (figure 24) are
diagnostic of Late Prehistoric I cultures (Prikryl, 1987: 133; Ferring and Yates, 1998).
Prikryl (1987) maintains that most projectiles are made of quartzite, while chert was
used more frequently to make arrows during the second half of the period. Signature
artifacts include smooth, flat-bottomed ceramic vessels; Gary dart points and the
appearance of small, expanding stem arrow points such as the Scallorn type and an
110
Figure 24. Late Prehistoric projectile point chronology for north-central Texas (compiled from Turner and Hester, 1993; illustrations used with permission).
111
assortment of tools including drills, scrapers, and grinding implements (Story,
1990:217, 356). Ceramics associated with Late Prehistoric I cultures are often tempered
with grog and bone (Brown and Lebo, 1990: 8). The most common cultural features
encountered are hearths and probable roasting ovens along with less frequently
occurring storage pits, burials, and possible structural remains (Story, 1990:356).
Limited evidence suggests that this period in north-central and east Texas evolved
directly out of the earlier cultures associated with the Late Archaic period. However, it
is important to consider the possibility of secondary information diffusion and recycling
that may have occurred due to site reoccupation. Dating of these technological changes
is speculative at best since a considerably earlier dating for the introduction of ceramics
has been suggested for portions of the middle Trinity drainage and for east Texas.
Late Prehistoric II occupations, in north-central Texas, are signified by Washita,
Harrell, Toyah, Maud, Perdiz, Steiner and Fresno arrow points (figure 24). Material
culture exhibits a distinct change in the composition of lithic tool assemblages. Other
material evidence includes bone tools; scrapers; edge-beveled knives; drills and shell-
tempered pottery (Prikryl, 1987; Brown and Lebo, 1990). Ceramics, projectile points,
and other material culture remains from archaeological contexts in north-central Texas
appear to be related to sites in eastern Texas from this same period (Story, 1981).
Introduction of the bow and arrow probably brought about a tremendous change in
hunting technique and efficiency, probably leading to a higher abundance of food
resources and a resultant increase in population. The shift from dart points to arrows
also implies adoption of individual hunting strategies over group hunting strategies.
112
Late Prehistoric Climatic Conditions
Around 1060 BP it is posited that the end of the West Fork Paleosol indicates a
return to a drier climate (Brown and Lebo, 1991:9). The presence of Paleosols between
2000BP and 1000 BP suggests a decrease in moisture during this period with a return
to wetter conditions after 1000 BP indicated by an increase in arboreal pollen (Skinner
and Ferring, 1999:10). The fairly steady decrease in the ratio of nonarboreal pollen to
arboreal pollen inferred from the Ferndale Bog Sample reveals a general trend toward
less dry conditions (Story, 1981). High grass pollen also occurs around approximately
1500 BP and during an interval between 500 to 400 years ago indicating that drier
climatic intervals probably interspersed with the wetter conditions that are suggested by
carbon isotope evidence (Ferring and Yates, 1998; Skinner and Ferring, 1999).
Late Prehistoric Subsistence Patterns
While changes appear to have occurred in most aspects of cultural lifeways, in the
Late Prehistoric, subsistence patterns changed little from Late Archaic times. In
general, evidence suggests that exploitation of riverine resources continues and the
exploitation of bison increases after A.D. 1200. However, throughout the Late
Prehistoric deer are considered to be the primary source of meat protein (Lynott, 1981,
Skinner and Baird, 1985; Ferring and Yates, 1997). Smaller game that supplemented
the meat portion of the diet predominantly includes rabbits and turtles. Plant foods
played an increasingly larger role in subsistence as indicated by the increase of
groundstone artifacts. Evidence from many sites in north-central Texas suggests that
the most common plant food source is the hickory nut (Martin, 1987) that would have
113
required grinding techniques in order to facilitate preparation for use. No tropical
cultigens have been identified in floral assemblages, although squash and sunflower
seeds are represented (Story, 1981: 146). Use of domesticates is indicated at several
sites (Peter and McGregor, 1988), but is not presumed to have been a dominant
economic focus. Corn horticulture prior to A.D. 1200 has only been found at a few
sites. As Story (1981: 146) proposes, subsistence was probably based on an intensive
foraging strategy and possibly included cultivation of native food resources. Ceramic
vessels are also an important innovation that may reflect alternative food preparation
techniques unknown during Archaic times (Story, 1981:146). This also may indicate the
beginning of an intensified subsistence strategy that may help explain reduction of
group mobility and population growth (Bruseth and Martin, 1987:284). In general, it is
hypothesized that Late Prehistoric cultures participated in a nomadic broad-spectrum
subsistence economy that included a primary focus on deer and seasonal scheduling.
The primary difference between the first and second phases of the Late Prehistoric
is related to climatic conditions. A more xeric climate is believed to have been a
dominant influence during the Late Prehistoric I era (Lebo and Brown, 1990). As such,
there is no indication of bison at this time. During the Late Prehistoric II period the
presence of bison remains at archaeological sites is considered to be supporting
evidence of the return to moister climatic conditions (Dillehay, 1974; Lynott, 1979;
Brown and Lebo, 1990). There is strong evidence from at least two archaeological sites
identified in Dallas and Denton counties that suggests a shift back to bison hunting and
a bison-oriented economy during the second phase of the Late Prehistoric time period
114
(Story, 1990:359; Ferring and Yates, 1997:6). One of these sites yielded two bison
scapula hoes and a tibia digging stick, while the other produced of eight bison scapula
hoes (Story, 1990:359; Ferring and Yates, 1997:6). Bison scapula hoes suggest a
combined economy of bison and rudimentary horticulture (Brown and Lebo, 1990).
Late Prehistoric Settlement Patterns
During the Late Prehistoric period site locations are very similar to the Late Archaic
in that they are situated along major tributaries such as the Elm Fork (Skinner and
Ferring, 1999:10). Sites are often found in the floodplains of area streams, especially
those of tributary streams. Strategic site selection is indicated in that sites were once
again located on sandy terraces above the floodplain in optimum settings.
Settlement patterns during the Late Prehistoric I era were effected by climatic
changes and technological adaptations. Important changes include shifts in settlement
patterns to larger sites that contain features including possible house structures.
Evidence such as the introduction of the bow and arrow suggests that there was an
external influence (Brown and Lebo, 1991: 9). More permanently settled villages are
expected related to the presence of maize. The transition from hunting and gathering
to a fully sedentary lifestyle was gradual in north-central Texas, probably beginning
during the Late Archaic but not established until the Late Prehistoric (Story, 1981:146).
Late Prehistoric II evidence suggests a shift in subsistence and settlement patterns
of the previously nomadic bison hunting groups. During this time period subsistence
became focused on riverine habitats with more sedentary settlements situated along
the major drainages (Brown and Lebo, 1991:9). Increased sedentism resulted in more
115
focal subsistence strategies (Brown and Lebo, 1991:9). These adaptations are thought
to be the result of internal population growth and external population pressure (Brown
and Lebo, 1991:9). Use of horticulture economies added another dimension to
subsistence opportunities (Brown and Lebo, 1991:9). Bison appear to have been
hunted on a more opportunistic basis rather than by the previous continual strategic
pursuit (Brown and Lebo, 1991: 9). Increased sedentism indicated by site size and
artifact assemblages is related to more focal subsistence strategies. Accordingly, in
north-central Texas, most of the Late Ceramic period sites, affiliated with the Caddo,
are concentrated at the eastern edge of the Eastern Cross Timbers. Over 80% of all
sites dating to this period along the Elm Fork of the Trinity are found at the prairie-
forest boundary (Prikryl, 1990:82). This concentration of settlement is hypothesized to
indicate responses to drier climatic conditions. “Neo-American” groups, who are
contemporaneous with Late Prehistoric cultures, followed a similar seasonal exploitation
pattern until agriculture became an important part of their subsistence base. In
general, during this time period the advent of agriculture led to the establishment of
large base camps or permanent villages that were typically located in or adjacent to
large standing alluvial bottoms of the more permanent water sources.
Unanswered Questions
Previous investigations have documented archaeological sites and formulated
theories concerning prehistoric settlement patterns in north-central Texas. Based on
these previous investigations it is possible to evaluate material culture, subsistence
strategies and settlement patterns as they change through time. It is also possible to
116
identify the gaps and questions that remain unanswered. While archaeological remains
are a known there are many unknowns that remain regarding how the people
interacted with the landscape and each other. Clues from prior investigations need to
be assembled into a coherent whole, gaps identified and new hypotheses tested, in
order to view the related behavioral patterns. Some of the unanswered questions can
only be answered by future research. In the meantime, the current research is
designed to synthesize and evaluate the results of previous investigations and formulate
hypotheses to fill in these gaps, where possible, answer some of these questions and
facilitate the construction of a regional model of prehistoric settlement pattern behavior
in north-central Texas. In order to accomplish this task it is necessary to construct a
unique methodology that has multi-scalar flexibility and a strong theoretical foundation.
117
CHAPTER 5
RESEARCH METHODOLOGY
Site catchment analysis (SCA) is an effective means to address hypotheses
formulated to answer the variety of questions posed by previous investigations and to
develop regional settlement models of prehistoric socio-economic behavior. Given the
multi-facetted character of the landscape, it is necessary to look for ecological
relationships that provide a strong economic resource base rather than one generalized
characteristic in isolation. This extends beyond physiographic settings, to include soil
interpretations, hydrological, geological and topographical relationships.
Ultimately the characteristics related to site location are the primary variables
correlated with site function, site type, occupation frequency, site interaction and
chronological relationships. The series of inter-connected relationships contribute to the
overall ecological structure. These relationships are evaluated to determine the
combinations of environmental features that influenced the choices and preferences of
prehistoric populations which are reflected by the corresponding settlement patterns.
The dynamic character of ecosystems requires that the inter-relationships of the various
components are evaluated separately in relation to a variety of factors and then joined
back together. Therefore it is necessary to test multiple hypotheses from a variety of
angles in order to gain a deeper understanding of how these components work together
to form a regional picture of prehistoric populations developing through time.
Essentially where a site is situated on the landscape is the primary geographical
foundation for SCA. Based on the location of each archaeological site and the character
118
of the artifact assemblage, a variety of research hypotheses are designed to evaluate
previous settlement pattern theories to answer the questions that prior investigations
left unanswered. In north-central Texas previous research has primarily evaluated site
location with respect to resource concentration in relation to physiographic settings,
hydrological patterns and topographic positioning.
For the current research, SCA is conducted on a regional scale to define the
character of 1km site catchment areas around archaeological sites. Sites are
collectively evaluated to formulate a regional model of prehistoric settlement patterns in
north-central Texas. SCA is performed by depiction, extraction, measurement and
evaluation of environmental data in correlation with archeological remains.
Environmental variables are derived from modern habitat descriptions. Data are
assembled on landforms (topographical, geological, hydrological) and potential of soil
types to produce economically advantageous (edible, medicinal, supplemental) plant
communities tabulated. Cultural variables are based on ethnographic analogy supported
by the evidence of the artifact assemblages collected from Texas Archaeological
Research Laboratory (TARL) files and a variety of reports. Criteria considered as viable
explanatory mechanisms for site selection and related settlement patterns include:
1. soil settings (taxonomic, textural, topographic) 2. soil potential to provide suitable habitat for plants and animals 3. economic value (edibility, medicinal, supplemental) of soils 4. catchment diversity 5. geological resources (surface geology) 6. geomorphology (uplands, bottomlands, floodplains) 7. physiographic resources (prairies and Cross Timbers) 8. climatic variables 9. cultural factors (archaeological composition and character)
119
Most of these criteria are mapped by utilizing digital data layers derived from several
sources. Some data layers did not exist prior to this research and had to be generated.
Conceptual criterion such as cultural factors, climatic variables and catchment diversity
are incorporated into the discussion rather than depicted on a map.
Pre-existing environmental digital data layers consist of County boundaries
(NCTCOG), geology (USGS), soils (SSURGO), streams (NCTCOG), Rivers (TNRIS), lakes
(NCTCOG), elevations (USGS) and physiographic regions (TNRIS). While the pre-
existing digital data layers are highly accurate it is important to note that there may be
slight inconsistencies, distortions or misalignment of the data layers, depending on the
scale of the data layer combined with the source that created the data. For the most
part, the data layers are seamless, consistent and accurate, especially those
constructed as a regional dataset by the same group of investigators. However, in the
case of the soils data layers there are slight inconsistencies in soil series classifications
that create seams at the County boundaries where proximal soils were categorized into
different soil series classifications. This is due to the fact that different investigators
conducted the soil surveys at different times and any slight categorical variation,
including utilization of a different name for the same soil type, will be enhanced when
the data is in digital format. In some cases it has to do with the level of detail and in
others the discretion of the particular soil surveyor. These differences are not
significant enough to affect the validity of the SCA but are brought to the reader’s
attention so that they may be aware of possible inconsistencies in the digital data
layers. This type of phenomena is taken into consideration in the current research and
120
should be addressed in any research that involves digital data layers created by
different sources.
Generated data layers include archaeological sites, site catchment boundaries,
economic values (edibility, medicinal, supplemental), topographic settings, riparian
corridors and ecotones. Environmental modeling of these data results in the mapping
of the distribution of topographic settings, physiographic settings, geological structure
and hydrological resources, plant and wildlife potentials and soil economic values.
While over the past tens of thousands of years there have been shifts from forests
to prairie environments in north-central Texas palynological and climatic evidence
indicates that over the past 8,000 years these changes are not extreme enough to
abandon inferences about prehistoric environments based on modern environmental
attributes. As such, modern soil types combined with their associated plant and animal
communities are a viable source for reconstructing prehistoric adaptive strategies. The
geological structure that forms the foundation for these soil types is stable enough to
support the development of similar soil types through time. What differs is the degree
of soil development and the climate that would have either facilitated abundant or
sparse resources related to those particular soil types.
A conceptual diagram (figure 25) provides an overview of the links and the data that
are collectively analyzed, in relation to archaeological site distribution, to shed light on
prehistoric settlement pattern behavior. The procedures used to perform SCA include
data collection, construction of databases, classification, scoring, data layer generation
and a variety of GIS techniques. GIS software also provides an interface to organize,
121
Figure 25. Conceptual diagram of data included in SCA.
Upland
Terrace
Floodplain
Geomorphology
Ecotones
Riparian Corridors
Woodlands
Grasslands
Soils Geographic Setting
Subsidiary
Medicinal
Edibility
Economic
Interpretations
Great Group
Soil Setting
Relationships
Rangeland
Wetland
Openland
Herbaceous
Grasses
Grains
Vegetation
Potential
Soils/Vegetation
Geology
Water
Supplemental
Medicine
Subsistence
Wildlife
Vegetation
Soils Potential
Climate
Geomorphology
Interpretations
Archaeology
Artifacts
Features
Choices
Prehistoric Settlement Patterns
Vegetation
Geology
Hydrology
Topography
Bottomland
Ridges
Wildlife
Site Catchment
Resources
Fauna
PhysiographyPhysiographic
Ephemeral Streams
122
manage, analyze and extract data to perform a variety of spatial and statistical
analyses. Terminology related to these GIS techniques is defined in a glossary provided
at the end of the appendix. GIS procedures are designed to conduct SCA and prepare
the data to address a variety of research hypotheses that collectively serve to build a
model of prehistoric settlement patterns.
Currently there are no known SCA studies that have set a precedent for the
particular approach created for this research. The current methodology is innovated to
respond to a series of questions regarding cultural affinities, contemporaneous
environmental opportunities, activities and subsistence strategies. Theories based on
past and present research are also tested to determine whether or not they are
applicable on a regional scale.
Soils as a Diagnostic Tool
SCA involves reconstructing the natural environment in past tense. The most
complex part of environmental reconstruction is associated with vegetation. In this
particular study soil is selected as a primary diagnostic attribute for the vegetation. As
a result a number of different investigators have employed a variety of techniques to
accomplish this goal including use of pollen diagrams, field methods, General Land
Office Surveys (GLO) and other historical surveys (Tiffany and Abbott, 1982: 314).
However, in this research, vegetation reconstruction is primarily based on soil potential
as inferred by soil survey data. According to Tiffany and Abbott (1982) the “systematic
changes in soil properties across a toposequence serve as proxy data for detailed
vegetative reconstruction” and “descriptions of soil types state the past vegetation
123
responsible for the formulation of that particular soil type” (Tiffany and Abbott, 1982:
316). Various SCA studies have employed soils as an explanatory mechanism. Soils are
utilized in the current research because they provide the foundation for the potential of
vegetation. Geology and soils are the most stable components of the environment.
Geology is the most stable as it is not subject to run-off and it is the parent material for
the soils. As such if there is increased rainfall that contributes to erosion the soil
formation process replenishes the ingredients of the soils that are subject to run off.
The character of the geology ultimately determines the character of the soil that
provides the matrix of possibilities for specific types of vegetation to exist. For
example, sandstone tends to produce sandy soils that lead to forest vegetation and
limestone tends to produce soils that are high in clay content creating an appropriate
environment for prairie vegetation. Any variation within the geology will lead to
variation within the soils which in turn leads to variation within the vegetation and this
is what gives the environment its diverse character. This variation is described by soil
taxonomy.
Climatic patterns in combination with topographic settings explain the abundance
and spatial distribution of specific types of vegetation that exist in response to the soils
and geology. Based on a sample of plant types, identified in the soil survey, research is
performed to determine plant edibility, medicinal value and supplemental value.
Economic potentials of soils are determined by soil survey correlations of soils to
vegetation according to settings where particular soils are typically situated. This
provides an interpretable view of potential prehistoric habitats. While the current
124
research is highly dependent on soils to ascertain economic value other environmental
variables are included to provide a more comprehensive view.
Soils Variables
Soil surveys contain an abundance of environmental information. Although many
descriptive attributes are attached to soil types the particular attributes that are used
for this study are limited to soil series, great groups, plant type potentials, wildlife
potentials, soil settings and topographic settings. Soil series are consolidated or
recoded to represent each of these variables. Pre-existing soils data layers exist for
each of the counties included in the regional study area (the West Fork or Wise County,
the Elm Fork or Denton County and the East Fork or Collin County). The only two
comparable soil surveys from the study area that are amenable to detailed analyses are
from Wise (West Fork) and Denton (Elm Fork) counties. Soils data for the East Fork of
the Trinity are limited to great group. As a result there are no equivalent correlations
with vegetation. Any relationships would have to be inferred. Although “great group”
scale soils data provide a means of comparison between the environmental contexts of
the East, Elm, and West Forks, the character of the environmental setting around the
East Fork is not suited for comparative analyses. The East Fork is clearly a significantly
different environmental setting than the Elm and West Fork of the Trinity. As such it
does not make sense to statistically evaluate the three settings comparatively.
Therefore, the East Fork is evaluated in a similar manner where possible and included in
comparative discussion but not statistically compared. Due to the lack of data currently
available statistical analyses involving the relative potential for vegetation types, wildlife
125
habitat and economic potential of site catchment areas are limited to the Elm Fork and
the West Fork of the Trinity.
Site catchment areas are composed of various scales of proportional values. Soils
data can be consolidated into soil series, order, settings or great groups. During the
organizational process of soils data it is important to note that a great group can consist
of several soil series or settings and likewise soil settings can include several great
groups. Soil series, however, are associated with one particular great group.
Therefore, the manner in which the data is consolidated is dependent on the research
question. For the current research the most appropriate approach is to conduct the
initial analysis on a soil series level and then consolidate the data into larger categories
when necessary. In order to recreate the variable potentials while considering soil
variability a soil series level approach is appropriate and preferable. This facilitates the
detection of variation within the great group while calculating plant potentials, wildlife
potentials, soil setting, topographic setting and catchment diversity. For the economic
analysis it is also necessary to work primarily from a soil series level and then
consolidate to soil settings and great groups to summarize the results.
Soil Settings
Soil series data is connected to attributes that describe taxonomic, topographic and
textural settings typically associated with certain soil series. These settings are
depicted by adjusting the symbology of pre-existing soils data layers in ArcMap.
Topographic settings associated with soil types include alluvial terraces,
bottomlands, floodplains, low hills and ridges, old terraces and valley fills, plains,
126
Pleistocene terraces, ridges, stream divides, stream terraces, terraces, upland ridges
and uplands. These settings are often referred to in the literature. While this provides a
sketch of the topographic setting the interpretive value of the particular setting is
increased when related to climatic conditions, physiographic settings, economic values,
related soils and faunal assemblages. Research conducted by Ferring and others
suggests that prehistoric populations primarily selected sites located in terraces and
upland settings in ecotonal transitional zones between prairie and woodland settings.
Hayes classifies sites on the Elm Fork into floodplain, terrace and upland settings.
According to previous investigations, during the Early Archaic, sites are situated on
stream terraces and in bottomland settings. Given the wet climate during the Early
Archaic it is expected that populations were concentrated in upland settings near
ephemeral streams. Given the dry climate during the Middle Archaic it is expected that
populations were concentrated in the bottomland, floodplain and riverine settings
associated with major tributaries. During the Late Archaic moister conditions are
expected to have increased the availability of plants for human consumption and
biomass for game animals. Vegetation retreated into the floodplain while drier
conditions prompted a return to the stream banks. Evidence also suggests utilization of
upland prairie settings. However, in general, subsistence strategies appear to have
been focused in riverine and bottomland settings within the prairies and Cross Timbers.
SCA techniques are designed to evaluate this combination of factors in relation to
occupation frequency and change through time. Patterns are also considered
comparatively between the Elm, East and West Forks of the Trinity. In order to gain
127
insight on these patterns hypothesis testing includes topographic settings in discrete
statistical analysis to evaluate site location with respect to occupation frequency and
change through time.
Soil textural settings include blackland, clay loam, clayey bottomland, claypan
prairie, deep Redland, deep sand, eroded blackland, loamy bottomland, loamy prairie,
loamy sand, low stony hill, Redland, rocky hill, sandstone, sandy loam, sandy, shallow
clay, shallow, steep adobe and tight sandy loam. These groups are primarily utilized in
the economic study to provide a linking mechanism between the vegetation and the soil
series. Great groups that are included in the study area (table 2) are evaluated in both
multivariate and discrete statistical analyses.
WEST FORK AND ELM FORK EAST FORKGREAT GROUP GREAT GROUP ARGIUSTOLLS CHROMUDERTS CALCIUSTOLLS CHROMUSTERTS HAPLUDERTS OCHRAQUALFS HAPLUSTALFS HAPLAQUOLLS HAPLUSTEPTS HAPLUSTOLLS HAPLUSTERTS PALEUSTALFS HAPLUSTOLLS PELLUSTERTS PALEUSTALFS RENDOLLS PALEUSTOLLS USTOCHREPTS RHODUSTALFS USTIFLUVENTS
Table 2. Soil taxonomy classifications. Soil series are consolidated into these great groups based on links established by the
soil survey. In order to attach economic value to these settings a “dissolve” operation is
performed rather than adjusting symbology. This allows the economic values to be
summed in relation to the consolidation of soil series into soils settings or great groups.
The relationship between the economic value and the particular soil setting is important
128
to the current multivariate analyses. For other parts of the analysis the archaeological
data and soil settings are evaluated to facilitate discrete analysis. In order to create
contingency tables ArcMap tools are used to select the particular settings by “attributes”
and archaeological sites (grouped into occupation frequency and according to
chronology) by “location” in relation to the particular selected settings.
Soil Potential for Plants and Wildlife
Plant types and wildlife potentials are associated with soil series based on soil
survey estimates within the overall study area and within each site catchment area.
These attributes are recoded (table 3) from soil survey ratings that were “high”,
“medium”, “low” and “very low”. Based on the frequency of “very low” categories for
these types of wildlife habitat, “low” and “very low” were combined. Soils that were
considered to have a “high” suitability for each type of habitat were given a value of
0.64. The soil types with “medium” potential were given a value of 0.29 and the soils
with “low” potential were given a value of 0.07. These weighting techniques are
derived from “Ranking Procedures” in GIS and Multicriteria Decision Analysis
(Malczewski, 1999:180).
Rank Sum Rank Reciprocal Rank Reciprocal Soil Rank Weight Normalized Reciprocal Normalized Weight Normalized
Suitability r = rank n = # of categories Weight Weight Weight Weight
r (n-rj+1) Weight/ Total Weight (1/rj)
(n-rj+1) p, where p=2
HIGH 1 3 0.50 1.00 0.55 9 0.64MEDIUM 2 2 0.33 0.50 0.27 4 0.29LOW 3 1 0.17 0.33 0.18 1 0.07Total 6 1.00 1.83 1.00 14 1
Table 3. Weighting procedure for plant and wildlife soil survey estimates.
129
Essentially this procedure transforms discrete into continuous variables so that they
may be proportionately assigned a value and evaluated in relation to their relative
representation within site catchment areas. These weights are attached to soil series
digital vector data by utilizing ArcMap to perform an “attribute join” operation. Once
the attribute data is joined to soil series location, with GIS tools, the spatial distribution
of these factors can be included in a multi-scalar spatial and statistical analysis.
Resulting data layers provide an overview of the spatial distribution of wildlife and
vegetation potentials for the entire County surfaces. In relation to archaeological sites
these data layers can be “clipped” based on the extent of site catchment areas with
ArcMap geoprocessing tools. Associated values that remain within the site catchment
territory can then be measured, described and statistically evaluated.
Interpretations: Potential Economic Value of Soils
Locally available botanical resources that could have been utilized by prehistoric
populations are of primary importance to SCA and settlement pattern investigations.
Although, ethnographic botanical information is sparse, a variety of publications have
been written that provide extensive information on the edibility, medicinal, and toxic
properties of plants that are native to Texas. In 1999, a comprehensive study
documenting the flora of north-central Texas compiled by Diggs, Lipscombe and
O’Kennon, was published. Tull (1987) put together a guide that discusses edible and
useful plants in Texas and the Southwest. Duke (2000, 2002) has written a number of
books that provide information about medicinal plants and herbs in the United States.
McCormick, Filson and Darden (1975); Skinner and Baird (1985); Lynott (1977); Martin
130
and Bruseth (1987); and others also provide information regarding the potential of
botanical resources in north-central Texas. According to Prikryl (1990), one of the best
publications concerning floral resources of the lower Elm Fork area is that of
McCormick, Filson and Darden (1975), who studied the terraces of an area where
Timber Creek enters the Elm Fork floodplain. Situated on the boundary of the Eastern
Cross Timbers and Blackland Prairie, this locale provides a healthy representative
sample of the vegetation composition in this area (Prikryl, 1990: 12). In addition to the
archaeological investigations a floral resources study was performed in immediate
vicinity of a major site. Of the 122 plant specimens collected, 97 were identified and 56
believed to have been utilized, for food, medicines, or technologically by the aboriginal
inhabitants of north-central Texas (McCormick et Al., 1975: 77). These resources are
utilized to aid in the interpretation of the botanical data retrieved from archaeological
sites and their site catchment areas.
For the purpose of the current research it was decided that vegetation would be the
primary determinant of soil economic value given its strength for interpretation and
correlation with soil types. Archaeological faunal data, although economically valuable,
is not incorporated into the determination of relative economic value of the soils even
though it contributes to the knowledge base of prehistoric choices. In the current
research economic value refers to the potential of a given soil type to provide
vegetation that would be economically valuable to prehistoric populations. The
relationship between certain soil types and specific plant types is an assumption that
has been empirically verified by soil scientists and can be tested by future research.
131
The lack of archaeological fauna in relation to all of the sample soil types and soil
settings would have created an inconsistency in the representation of economic values.
The relationship of archaeological fauna to soil type is also less predictable although
future research could test relationships between soil type and archaeological fauna to
try to predict where comparable fauna and their corresponding prehistoric populations
might be located. Faunal data are incorporated into the “overall” economic value and
soils portion of the analysis by mapping the potential for specific wildlife types, in
relation to their habitat, associated with soil types. During the final synthesis the
recoded soils categories, including potential for wildlife habitat, are combined to
determine the overall economic value. However, the correlation between soil type and
economically valuable vegetation is the strongest and provides the highest resolution.
Mapping soils data across a region is one of the most detailed views available of the
environmental character of a particular landscape. Vegetation data, correlated with
soils, sheds light on the potential of the particular types of soils. When the vegetation
is interpreted into use-categories, as related to human populations, the soils provide
culturally significant information. Economic values were determined by evaluating the
botanical, ethnographic and archaeological literature to determine if the sample
vegetation, associated with soil types, was edible, medicinal or supplemental. Edible
vegetation is defined by whether or not the vegetation could be ingested, if it has
served as a food resource for prehistoric or modern populations and the likelihood that
it would be utilized for food based on the relative caloric value. Medicinal vegetation
was defined by the known use of particular plants to be used to cure internal ailments
132
or topical wounds. Many of these plants are highly toxic and require a certain degree of
preparation. Therefore, it was decided not to factor in the toxicity of plants as a
negative impact on whether or not the plant was used by prehistoric populations.
Supplemental vegetation primarily includes wood for fires or construction and plants
that would have been utilized for dyes, basket making, or various types of construction.
Codes are assigned to a sample size of 242 vegetation species collected from
associations made in the soil surveys and archaeological literature. Based on the
edibility, medicinal value, and supplemental potential of the plant the relative values are
translated into economic values per category. The resulting scores are arbitrarily
calculated by assigning a percentage to the applicable category. For example, if a plant
has both medicinal and edible properties each category is given a value of 50 percent.
If a plant is primarily edible with limited medicinal value edibility would be given a value
of 90% and medicinal properties a value of 10 percent. Each percentage is determined
based on the information provided in the descriptive literature. Often a several
resources were evaluated to determine the edible, medicinal and supplemental
properties of the plant in order to take into consideration the likelihood that these
particular plants would have actually been utilized by prehistoric populations. The
resulting scores are provided in the appendices.
Plants are associated with the appropriate soil setting and resulting economic values
are related to soil series. Therefore the 242 vegetation species in the sample are
consolidated into 22 soil settings. The general procedure followed to assign economic
meaning to the landscape is summarized by a conceptual diagram (figure 26). The
133
economic values of the soil settings are linked to the soil series and are proportionally
evaluated within site catchment areas. Ultimately, soil series are consolidated into the
11 great groups that are represented in the site catchment areas so that they may be
summarized. Consolidation of the appropriate soils series provides a relative economic
value for the great groups within each catchment. These values are summed and
assigned to the catchment area to provide an overall economic value. Each catchment
area is also assigned three other thematic economic values (medicinal, edibility,
supplemental) by establishing the same data links. Proportional distribution of the soil
series within the catchment areas is constant. What changes are the economic values
associated with the soil series. This is how the relative edibility, medicinal,
supplemental and economic values are measured in relation to each site catchment.
Economic values and potentials for specific wildlife and vegetation habitat that are
assigned to soils help interpret resource concentration patterns. It has been posited
that sites are primarily located near a major source of food. SCA can shed light on the
estimated economic value of such settings, the kinds of food or environmental
resources that were accessible in the immediate vicinity of site locations and answer
questions related to specialization, diversity and optimization. Furthermore, resource
availability in correlation with site location can shed light on the theories regarding the
use of this region for seasonal migration as suggested by McCormick. Skinner and Baird
hypothesize specialized ecological approaches in the northern reaches of the Elm Fork
correlated with high yield mast crops in the uplands and central locations of the Cross
Timbers. McCormick (1975) noted that sites along the southern reaches of the Elm Fork
134
GREAT GROUP ARGIUSTOLLS CALCIUSTOLLS HAPLUDERTS HAPLUSTALFS HAPLUSTEPTS HAPLUSTERTS HAPLUSTOLLS PALEUSTALFS PALEUSTOLLS RHODUSTALFS USTIFLUVENTS WATER
Site Catchment
Area 1 km2
Soil Series
SOIL SETTING BLACKLAND CLAY LOAM CLAYEY BOTTOMLAND CLAYPAN PRAIRIE DEEP REDLAND DEEP SAND ERODED BLACKLAND LOAMY BOTTOMLAND LOAMY PRAIRIE LOAMY SAND LOW STONY HILL REDLAND ROCKY HILL SANDSTONE SANDY LOAM SANDY SHALLOW CLAY SHALLOW STEEP ADOBE TIGHT SANDY LOAM WATER
SOILS POTENTIAL FOR VEGETATION
POTENTIAL FOR WILDLIFE
ECONOMIC VALUE EDIBILITY
MEDICINAL
SUBSIDIARY
VEGETATIONSOIL SERIES
Figure 26. Procedure to link economic value to site catchment.
are primarily located along the upland edge where maximized resource zones are
accessible. Riparian fauna are associated with archaeological remains along the Elm
Fork indicating concentrations in riverine settings. Sites located along the East Fork
are unquestionably situated in the Blackland Prairie. However, West Fork sites have not
been previously evaluated in prior investigations in relation to their respective
135
physiographic settings. Observation of site patterning along the Elm Fork in correlation
with these settings, on a larger scale, sheds light on regional behavior. Given that
information about these sites is primarily derived from state files there are many
unanswered questions. SCA facilitates the evaluation of site placement within particular
physiographic settings and comparisons of these locations according to occupation
frequency, density and change through time.
Economic values are represented in a series of analytical maps by recoding
attributes associated with the pre-existing soils data layers. These values are
associated with archaeological sites with respect to time period, occupation frequency
and site contents. Then they are statistically evaluated to test a series of hypotheses.
Site Catchment Soils Diversity
Soil series associated with sites provide the best resolution in terms of variation
within a catchment area. Each catchment area contains a square kilometer of soil
series. Each great group contains proportional distribution of soil series relative to the
location of the site. Therefore, the diversity is more accurately expressed by the soil
series. In order to gain a deeper understanding of the diversity within site catchments
each catchment is evaluated with regard to the proportional representation of each soil
series within each site catchment to calculate diversity. Environmental diversity indices
for site catchments measure the potential of an area for the presence of archaeological
sites based on the premise that site location is associated with increased diversity
(Tiffany and Abbott, 1982). For the current research an environmental diversity index
is developed that differs from previous studies in that it is based on the soil proportions
136
of soil types within site catchment boundaries. Given that all of the sites are in the
vicinity of water and that there is a strong correlation between geology, soils and
vegetation a different technique was chosen. This technique is referred to as the
Shannon-Weiner Diversity Index (Colinvaux, 1993: 316),
H1=-Σ pilog2pi,
where pi= relative abundance of soil type i to total soil, is employed to evaluate the
relative diversity of each catchment. By creating an environmental diversity index
based on proportional soil types within the site catchments the current research seeks
to evaluate if there is a correlation between environmental diversity and prehistoric
preferences for selection of site location in north-central Texas.
Site catchment diversity is evaluated in relation to occupation frequency and change
through time in order to determine how variable diversity is at sites that contain
evidence of single versus multiple occupations and how diversity changes through time
in relation to site location. The diversity values in relation to these factors are analyzed
with an ANOVA statistical test.
Geological Variables
Geological formations provide the parent material and foundation for the range of
soils that can develop in a certain location. Given the minor differences in the climatic
regime of this region the majority of variation is dependent on the character of the
bedrock lithology. The geological structures (table 4) that are associated with
archaeological site catchments in the region provide a means to view similarities and
differences in more generalized site settings. This component influences the
137
character of the physiographic setting and is evaluated with respect to change
through time and occupation frequency in a discrete statistical analysis. The West Fork
and Elm Fork of the Trinity are generally associated with similar geological sequences.
Geological formations associated with the East Fork are notably different in character as
listed in table 4. While there is a strong correlation between soils and geological
settings geology is also culturally significant in that it serves a source for lithic
procurement. In addition, it provides a means to cross-check the results of the soils
analysis and to explain the particular variation of correlated soil types. Statistically, the
geology is tested in a different manner than the soils with a Chi-Square analysis rather
than with an ANOVA. As such, geology is evaluated independently of soils even though
they are inter-related.
WEST FORK AND ELM FORK EAST FORKALLUVIUM ALLUVIUM ANTLERS SAND AUSTIN GROUP BOKCHITO FORMATION (UNDIVIDED) FLUVATILE TERRACE DEPOSITS EAGLE FORD FORMATION MARLBROOK MARL FLUVATILE TERRACE DEPOSITS OZAN FORMATION GLEN ROSE LIMESTONE PECAN GAP CHALK GOODLAND LIMESTONE/WALNUT CLAY (UNDIVIDED) WOLFE CITY SAND GRAYSON MARL AND MAIN STREET LIMESTONE KIAMICHI FORMATION PALUXY SAND PAWPAW FORMATION SURFICIAL DEPOSITS (UNDIVIDED) TWIN MOUNTAINS FORMATION WILLOW POINT FORMATION WOODBINE FORMATION Table 4. Geological formations associated with site catchment areas in study area.
138
Physiographic Variables
Physiographic variables that are evaluated in relation to archaeological sites include
the Western Cross Timbers, Grand Prairie/Fort Worth Prairie, Eastern Cross Timbers,
Blackland Prairie, riparian corridors, ephemeral riverine settings and ecotonal settings.
The “Buffer Wizard” in ArcMap is utilized to generate transitional zones around water
resources and physiographic boundaries. Ecotones are generated in ArcMap by creating
2 km buffers around physiographic boundaries. Riparian corridors are generated in
ArcMap by creating 2km buffers around major and minor tributaries in the streams
data. Ephemeral streams are also included but relative to their influence on the
landscape and intermittent character 1km buffers are constructed around these areas.
In the case of the current research physiographic settings include hydrological
resources because these areas are considered to be unique microenvironments.
Physiographic settings are often referred to in the literature in relation to site
location. While this is informative the interpretive value of physiographic settings can
be enhanced with data regarding economic value, faunal assemblage, vegetation
habitat, wildlife habitat inferred from the soils which are related to these physiographic
settings. In addition, larger patterns that emerge in relation to site function, occupation
frequency and change through time can shed light on prehistoric settlement pattern
behavior. A variety of investigators report that grassland, woodland, riverine and
ecotonal settings were utilized by prehistoric populations. Hypothesis testing sheds
light on more specific patterns that emerge in relation to these settings. It is likely that
139
given the various climatic conditions, availability of seasonal resources, and socio-
cultural conditions that certain areas were occupied more intensively than others.
If sites are situated primarily within grassland settings prehistoric populations would
have access to a high variety of grasses, legumes, grains, seed crops, rangeland and
openland wildlife. In addition, site selection along major tributaries would ensure the
availability of supplemental economic resources such as wood for cooking. If the sites
are situated primarily within woodland and riverine settings they would have access to a
high variety of wild herbaceous plants, fish, wetland wildlife and various smaller fauna.
In addition, woodland settings provide a greater abundance of supplemental material
such as wood. Woodland settings also provide access to an increased supply of nuts
and berries. It is expected that the potential for various types of vegetation within
these areas reflects an economically advantageous positioning. Given that both
grassland and woodland settings have a riverine component other factors must be
correlated.
If the pattern detected in the Lewisville report (Ferring and Yates, 1998) that
“prehistoric populations exploited whatever was available to them within well-defined
territories” is true then statistical analyses should reveal that a random effect is
operating. However there may be a pattern that reveals differential utilization of the
landscape. If clearly defined territories exist it is possible to infer the degree of social
influence on these boundaries in relation to site distribution and economic values. If
these populations were primarily effected by socially defined territories economic values
associated with site catchments should reveal that site selection was random in these
140
locations. Further, it is possible to evaluate the environmental variables contained by
site catchments and consider the possibility of environmentally defined territories.
Sites reported in the Lewisville report (Ferring and Yates, 1998) are situated
primarily in the ecotone between the Eastern Cross Timbers and the Blackland Prairie.
Sites in the Ray Roberts area are reported to be centrally concentrated in the Eastern
Cross Timbers. It has been suggested that sites are situated to equally utilize forest,
riparian and prairie environments. If it is true that these environments were equally
utilized then all of these types of soils should be within site catchments. Hypothesis
testing provides a means to determine if similar sites are situated in similar settings.
SCA can help determine what type of site is located in particular settings or where
equivalent sites are located. Sites should reflect the setting through faunal remains
and contents preserved by the archaeological record. Further, it is expected that
economic values of vegetation and wildlife within catchment areas support conclusions.
Physiographic settings are a dynamic variable that interacts with other physical
environmental factors. As such this generalized setting must be evaluated from a
variety of micro-environmental perspectives.
Physiographic settings reflect the inter-relationships between all of the components
that have been forming the modern landscape for millennia. A wealth of information is
related to the available faunal and floral communities contained by these settings. This
particular variable is tested for significance between site catchment areas with a Chi-
Square statistical test. In this case, the goal is to find out if certain physiographic
locations were preferred over others.
141
Climatic Variables
Climatic data available for past environments is sparse but correlations have been
made based on palynological data collected from peripheral locales. Pollen types are
selected from the Ferndale Bog sample only if they are represented in the vegetation
sample utilized for this research. This vegetation is correlated with soil series and
therefore can be depicted on the map. Vegetation and pollen being both associated
with “Plant Family” provided a means to link the results from the Ferndale Bog sample
data (figure 27) to the digital soils data layers and the correlated economic values
associated with site catchment areas. Prior to linking these data to the soil settings
they were calculated to take into consideration the relative representation of these plant
types within site catchment areas. In order to preserve the raw economic values the
pollen data is added to the original values associated with the same settings and
adjusted to reflect the change through time by building economic values for each time
period. These values are depicted a series of maps and correlated with sites in order to
evaluate if sites are situated in areas that are expected to be relatively abundant during
certain climatic intervals. The raw data values and the series of maps that depict these
associations per time period are included in the appendix.
Dawson and Sullivan propose that the character of the environment where
prehistoric sites are located is such that it was necessary for prehistoric groups to
remain small and scattered in order to maintain balance with minimum cycles
associated with climatic and seasonal fluctuations. This type of theory can be tested by
evaluating the correlations between occupation density, economic values and climatic
142
data. Site selection as related to topographic setting can be further enhanced by
incorporating climatic data. For instance, it is suggested that the terrace and upland
settings represent site selection and occupation during wetter climatic episodes. This
inference is tested by relating site locations to the fluctuations of pollen abundance data
based on the Ferndale bog sample. Hypotheses testing can determine if site locations
as they shift and change through time corresponds with climatic conditions.
SCA can reveal if site positioning is a deliberate selection that met the needs of
these populations given the character of the landscape combined with climatic data. If
so, SCA can help ascertain if there is any additional information that facilitates increased
understanding of these choices. If not, such studies can generate evidence to provide
alternative theories.
0% 20% 40% 60% 80% 100%
ACERACEAE
ASTERACEAE
CYPERACEAE
FABACEAE
POACEAE
SALICACEAE
TYPHA
ULMACEAE
EA (WET) MA (DRY) LA (WET) LPI (DRY) LPII (WET)
Figure 27. Relative Abundance of Plant Family in Study Area.
143
Archaeological Variables
Archaeological site coordinates were collected from TARL records, converted to
points in ArcCatalog and projected with ArcToolbox. Site catchment boundaries are
established in ArcMap with the “Buffer Wizard”. Although it is possible to construct time
contours with GIS techniques, the character of the landscape in the study area did not
provide a return on investment equal to the time it took to create the contours.
Therefore, the site catchment area is established for the 150 sites in the study area by
performing a buffering technique to construct an arbitrary 1-kilometer radius around
each archaeological site in the study area. Site catchment areas are constructed for 50
sites that contain diagnostic material and 100 surface sites that do not contain time
diagnostic material. The diagnostic sites are valuable in that they provide clues as to
the specific time period and duration of occupation. The non-diagnostic sites are
valuable with respect to providing evidence of prehistoric activities in relation to
location. When these sites are associated with the specific characteristics of the site
catchment areas they provide fruitful behavioral information in relation to the time
diagnostic sites. One could hypothesize that the sites containing lithic scatter are
ephemeral hunting, gathering, eating or resting spots. Whatever their purpose might
be they provide valuable comparative information in the fine scale analysis of
vegetation and soils data related to site catchment areas. Proximity to time diagnostic
sites also provides an informative view of the possibilities of inter-relationships.
Archaeological site data is categorized into groups to test for statistical significance
between site contents and catchments with respect to environmental variables,
144
occupation frequency and change through time. Activity groups include: tool
manufacture, food preparation, hunting and lithic scatter. Site files are evaluated to
infer activities. Lithic manufacturing activities are inferred if the artifact assemblage at
a site contains cores, hammerstones, performs, bifaces, debitage or a variety therein.
Evidence for food preparation includes tools, groundstone, ceramics, animal bones and
fire cracked rock. Any variety of these items is sufficient for the site to be listed as
including food preparatory activities. Hunting activities are inferred by the presence of
diagnostic projectile points. A separate category of lithic scatter is created to account
for sites where no functional diagnostic information is available. There are enough
cases where various densities of isolated lithic scatter are reported to prompt the
inclusion of this category. An increase in the number of activities represented by tool
types at a site indicates an increase in artifact diversity. It is a basic premise of SCA
that site function and site location are correlated and that inferences can be made
about function from knowledge of location (Roper, 1979:121). Based on previous
studies it is possible to form expectations regarding site function. For instance, sites
that contain a substantial amount of debris have often been classified as “home base”
sites, whether they were occupied year round or only during certain seasons. Previous
investigations report specialized activity sites such as “manufacturing stations” have
been located in the Cross Timbers. Hayes references the archaeology and suggests
that sites on the floodplain represent characteristic mussel procurement sites
referenced by Skinner. He does not specifically reference Skinner’s work but makes an
association between the faunal and artifactual remains, possible activities and
145
topographic settings. Hayes notes that terrace sites consist mainly of lithic scatter and
upland sites are found in areas where lithic raw materials may have been procured.
During the Middle Archaic projectile point size and diversity increases while mobility
generally decreases. It is expected that this type of phenomena is reflected in site
function and location. These types of relationships are investigated and enhanced by
environmental information. Hypothesis testing evaluates site types in other localities to
determine if they are concentrated or randomly distributed. McCormick et al. (1975)
reference similarity of activity at site locations regardless of time period. Factoring in
change through time will shed light on these types of temporal considerations. The
current research correlates site contents with the catchment to determine site function.
Site activities are initially inferred based on the character of the relative artifact and
faunal assemblages. This is intended to provide a simplified means to relate the site
contents to the site catchment. After hypotheses testing site function is verified or
revised.
Occupation density includes ephemeral, moderate and intensive categories. These
categories are formulated on the basis of the number of artifacts reported at a site.
Ephemeral occupation is defined by sites that contain one diagnostic artifact or limited
lithic scatter. Moderate occupation is defined by sites that contain more than one but
less than ten diagnostic artifacts and/or moderate lithic scatter. Intensive occupation is
determined based on more than ten diagnostic artifacts which are often found in
combination with more intensive lithic scatter and a variety of tool types. Previous
reports often reference the density of archaeological content at a site. A model based
146
on correlations with migratory patterns of bison rests on the assumption that these
prehistoric populations were following the herds. Sites related to this pattern are
expected to be more ephemeral in character. SCA facilitates testing for the presence of
temporary camps in relation to seasonal resources and if these sites are fewer in
number than during the Archaic as suggested by McCormick et al. (1975). It is
anticipated that during the Late Prehistoric time period temporary hunting camps will be
found in association with more sedentary base camps due to the opportunities that bow
and arrow technology offers. McCormick also posits that the zone of movement was
along the prairie-cross Timbers boundaries where sites are considered to be “more
permanent”. Sites reported to have the largest collections of Late Prehistoric I artifacts
are located on the Cross Timbers-prairie boundary. Ephemeral sites in some areas are
related to seasonal occupation that focused on harvesting the mast crop resources in
the region. Hypotheses regarding site function and occupation density can test for the
presence of permanent village, temporary campsites and activity-specific sites in the
study area. Occupation density is more meaningful when associated with
environmental attributes present at site locations.
Occupation frequency refers to how many times a site was occupied based on the
number of time periods represented at the site by diagnostic tools or radiocarbon dates.
The categories for occupation frequency include single, dual, tri and multi. Some
investigators have suggested that north-central Texas is a “marginal” setting and others
that the grassland, woodland and riverine settings are “highly opportune”. In order to
evaluate this facet economic value is assigned to soils to identify areas that provide a
147
more marginal or opportune setting. It is possible that during certain time periods the
region provided a marginal setting while during other time periods it provided an
opportune setting. Site density provides a means to evaluate where prehistoric
populations were focusing their efforts during these time periods. If certain locations
provided resources with a relatively higher economic value it is likely that these areas
contain evidence of multiple occupations. Economic values and site density in relation
to change through time provide the basis to test hypotheses that have been formulated
to answer these questions.
Younger surface sites often contain cultural materials of older time periods. This
creates a complex situation that makes it difficult to ascertain if these sites were
occupied by cultures from different time periods or if they represent one culture that
was reusing materials left behind by previous occupants of the area. To add to the
complexity it is unknown if these points were being recycled on location or if they were
collected from other locations and brought back to a central location. It is posited that
identical projectile point types were made by the same individuals or those who shared
technological knowledge. Therefore, it follows that these same individuals would have
also shared knowledge regarding effective plant utilization, hunting techniques and
paired with information regarding suitable locales to exercise these subsistence
strategies. As such it is possible that site catchments that contain the same projectile
point types are either similar in character suggesting that these individuals preferred
certain settings, or very different in character indicating resource concentration in
association with specialized activities and in either case exhibiting a relatively high or
148
low diversity. Hypotheses designed to test for significance evaluate site catchments in
relation to projectile point type.
Established relationships between projectile points and time periods provide the
temporal foundation for the current research. These data are the primary basis for the
evaluation of change through time. Given the types of projectiles found in north-central
Texas research hypotheses look for correlations between point types and environmental
features found in catchments. It is expected that similar point types would be found in
similar settings. If these point types are contemporaneous it is expected that the
people who utilized these tools not only shared technological knowledge but
environmental knowledge. As such it is expected that the prehistoric populations that
utilized these particular point types employed similar subsistence strategies. Given that
the environment would have offered a range of similar optimal opportunities during the
associated time period it is expected that environmental characteristics in association
with similar point types will reflect these opportunities. However, it is also possible that
given seasonal variation strategies will reflect the necessary adaptations. To evaluate
this facet a hypothesis is tested for statistically significant differences between site
catchments in relation to projectile point types. The sites are evaluated to determine
the range of settings that were being utilized during these time periods. Tool kits will
also be evaluated with respect to composition and occupation density to evaluate
activities.
Prikryl (1990) reports that there is a strong cultural continuity between the Late
Archaic and the Late Prehistoric at sites situated on the Elm Fork of the Trinity. Of the
149
36 sites containing Late Prehistoric I artifacts, 34 also contain Late Archaic tools.
Seventeen of the 34 sites were first occupied during the Late Archaic. Similar to the
Late Archaic, 35% of the Late Prehistoric I assemblages are recovered from sites whose
closest water source is the Elm Fork, while the other 65% are recovered from sites
whose closest water source is a tributary stream. Late Archaic settlement patterns
indicate increasing evidence for the reuse of camp-sites and a trend toward seasonal
aggregation of populations when large quantities of food were locally available (Lynott,
1981: 105). Patterns indicate that certain locations were preferred over time (Skinner,
1981) but the nature of these locations is not clear. SCA provides a means to define
these preferences and determine if sites contain evidence of repeated occupation
concentrated in one particular physiographic setting or multiple physiographic settings.
Hypothesis testing identifies favored environments and preferred settings and is
expected to reveal patterns that explain occupation frequency.
Change through time is determined based on the time periods represented by
diagnostic point types recovered from the site. The time periods that are evaluated
during the course of this research include Early (8,500-6,000 BP), Middle (6,000-3,500
BP) and Late Archaic (3,500-1,250 BP), Late Prehistoric I (1,250-750) and Late
Prehistoric II (750-350 BP). Sites are assigned to these time periods based on a variety
of site files, contract reports and academic publications. Evaluating change through
time in relation to site location provides a dynamic view of the interplay of these
variables that respond to climatic and socio-cultural conditions. As such, change
through time is an important facet that can address theories related to site positioning.
150
Correlation of site function and density with site catchment characteristics and change
through time can test theories that have been proposed regarding prehistoric
settlement patterns.
Topographic settings take on enhanced meaning when associated with change
through time. Dawson and Sullivan propose that Archaic sites in the East Fork situated
on ridges and Late Prehistoric sites are situated primarily in bottomland areas. If such
theories are reliable hypothesis testing should provide support or reveal differential
patterning that may be related to other factors.
Physiographic settings are expected to change in response to climatic conditions and
thus exhibit change through time. McCormick et al. (1975) claim that temporary
campsites occupy similar ecological niches and that this remains the case in both
Archaic and “Neo-American” sites. McCormick et al. conclude that Archaic seasonal
camps are located in eco-tonal settings and secondary activity-specific sites situated
with respect to “specific outcroppings, plants and water”. If Archaic camps are
primarily located in eco-tones SCA can help determine the “specific outcroppings,
plants, and water” that they are referring to. Hypothesis testing with respect to change
through time can also test whether or not Late Prehistoric populations follow similar
patterns as McCormick et al. (1975) suggested.
During the Early Archaic, cultural responses to new ecological systems include a
wider variety of projectile points, more variable artifact assemblages, decreased
mobility, seasonal scheduling, and changes in food processing. Subsistence patterns
involve a wider range of plant collection. During the Middle Archaic evidence suggests
151
that a wide variety of vegetation continued to be utilized paired with an increase in
openland economies. Although the importance of hunting appears to be generally
reduced throughout the Late Archaic evidence suggests the continuation of hunting in
the upland prairie. Utilization of a mix of prairie, forest and riparian species is also
indicated during the Late Archaic. Hypothesis testing with respect to site catchment
diversity, economic values and soil potential for wildlife and vegetation should reveal
such patterns or provide alternative perspectives. The added dimension of change
through time delineates types and varieties of resources utilized by prehistoric
populations of different time periods. SCA facilitates evaluation of site placement within
their respective physiographic settings and provides a means to compare these
locations to site content, occupation frequency and density.
A detailed analysis of 100 of the sites, along the Elm Fork and the West Fork of the
Trinity, is performed to evaluate the pattern of site distribution, land use, and possible
function, based on the assemblages and catchments of each site. The purpose of this
portion of the study is to analyze the catchments of diagnostic and non-diagnostic sites
for comparative analysis. The 45 time-diagnostic sites provide a means to observe site
characteristics with respect to change through time and occupation frequency. Since
the East Fork provides a clearly significantly different environmental setting than the
Elm and West Fork the 50 site catchments from this area are included in comparative
discussion but not in statistical analysis. Mapping the economic value of the entire
landscape reveals the overall distribution of valuable resources as they relate to site
location. This can help explain why certain areas were selected over other areas or
152
provide a means to predict likely site locations. Once the intra-site settlement patterns
are evaluated these data can be compared to non-surveyed areas of the Upper Trinity
drainage basin in order to ascertain the kinds of environments associated with the
various prehistoric utilizations of the basin.
The character of investigations in the region introduces a sample bias that
influences the results of the current research. The Elm Fork bridges Ray Roberts Lake
and Lake Lewisville. Professional contract work associated with the construction of
these reservoirs has led to the discovery of a very dense distribution of sites along this
corridor. Academic work conducted at the University of North Texas and Southern
Methodist University has also produced a number of reports along the Elm Fork. Site
distribution in the West Fork portion of the study area is very limited. This is due to the
fact that most of the sites were discovered by amateur archaeologists or land owners.
No formal professional investigations and no contract work, related to reservoir
construction, have been conducted in this portion of the study area. East Fork sites
have been discovered by limited excavation efforts related to academic research. Even
though distribution currently reflects the intensity of archaeological contract work, it is
clear that the Elm Fork found favor with prehistoric populations. Future professional
research should be conducted along the West and East Fork to further explore the
benefits offered in these areas and to substantiate or refute proposed settlement
models. Regardless of sample bias, the current sample provides valuable information
from which to form future hypotheses and conduct research on the settlement patterns
of prehistoric populations in north-central Texas.
153
Site Catchment Analysis of Prehistoric Sites in North-Central Texas
Geographic Information Systems (GIS) permits archaeologists to conduct
sophisticated spatial analyses by providing an extremely useful methodological structure
through which quantitative approaches to environmental attributes and cultural remains
can be spatially addressed. SCA is performed with ESRI ArcView 8.3 ArcMap and SPLUS
2000 statistical software. GIS techniques (figure 28) are an active component,
throughout the entire process of SCA, from data collection to hypothesis evaluation.
GIS provides an ideal set of tools to facilitate the process of performing SCA. The ability
of a GIS to extract environmental information and perform spatial analysis has led to its
utility in SCA. GIS software plays a central role in data collection, input, management
and analysis within site catchments.
Essentially the GIS component provides an organizational mechanism (figure 29) to
manage large data sets from a variety of sources, generate new data sets and integrate
these data to view the dynamic relationships that help explain prehistoric human
behavior. GIS software provides tools to facilitate the construction of databases and a
means to link the archaeological record with the environment, on a regional scale. It is
with these tools SCA is performed and with ecological, anthropological and geographical
theories that settlement patterns are interpreted.
154
DATA
COLLECTION
DATA INPUT
DATA
MANAGEMENT
DATA
ANALYSIS
RESEARCH
HYPOTHESES
OBJECTIVES
MAPPED
INFORMATION
DATA OUTPUT
DECISION
MAKING
ANALOG
DIGITAL
IMPORT
GENERATION
SPATIAL
ATTRIBUTE
SPATIAL
ATTRIBUTE
REPORTS, FILES, MAPS
GIS DATA LAYERS
DATA CONVERSION
RECODING
BUFFERED SELECTION
DATA EXTRACTION
OVERLAY ANALYSIS
POST-PROCESSING
Figure 28. Diagram of GIS data processing design in SCA.
155
GIS Archaeological Sites Archaeological Data
Spatial Locations Economic Data
Statistical and Spatial Analyses
Environmental DataSite Catchment (1km)
Cultural Reconstruction
Cultural Ecology
Archival Information
Archaeological Preservation
Environmental Reconstruction
Model of Regional Prehistoric Settlement
Pattern Behavior
Figure 29. GIS in relation to components of the settlement model.
156
In general, the SCA, which is a synthesis of archaeological, environmental and
ethnographic data, includes the following procedures:
1. Soil survey soil series level data are evaluated with respect to vegetation.
2. Vegetation, associated with specific soil types in the soil surveys are identified by common name, family, genus and species.
3. Research is conducted to determine the relative values, per plant species, of
edibility, medicinal value and supplemental value.
4. Values are attached to soil series level data in ArcMap GIS interface to the digital soils data layer.
5. Soil series are recoded with economic values, vegetation and wildlife potentials
using ArcMap “joining” operations to relate to site distribution in light of chronology, site frequency and artifact density.
6. Archaeological sites are projected with ArcCatalog and ArcToolbox and then
imported into the ArcMap GIS interface where 1 km catchment areas are constructed around the sites with ArcMap buffering techniques.
7. Soil types within catchments are proportionately extracted, recorded and
interpreted into measures of diversity, plant and wildlife potential and economic worth with ArcMap selection and extraction techniques.
8. Overlay analysis is performed using ArcMap to provide a visual representation of
the intersecting data layers that form the environment in association with sites and facilitate evaluation of spatial patterning.
9. Observable topographical, geological, hydrological, physiographic and ecotonal
resources are related to sites chronologically, with respect to occupation frequency and assemblage composition, by selection and extraction ArcMap techniques to compile contingency tables for chi-square analyses.
10. Descriptive and multivariate statistical tests are performed with SPLUS 2000
software to objectively compare the economic values, wildlife potentials, vegetation potentials, topographic settings, great groups, geological and physiographic settings within site catchment areas to look for patterns in site occupation, composition and change through time.
157
Results generated from the SCA are, in part, presented in a series of maps. A series of
thematic databases, charts, histograms, and graphs summarize the proportional
distribution of soil types, geological settings, topographic settings, physiographic
settings, economic potentials and catchment soil diversity. Site catchments are
summarized with descriptive and multivariate statistical techniques to describe, quantify
and compare the resource potentials of site territories. Based on the prior research a
series of hypotheses are tested and a model of prehistoric settlement patterns is
formulated.
Hypotheses Testing
In order to understand prehistoric human socio-economic settlement patterns in
north-central Texas this thesis specifically aims to formulate a model that can help
explain why prehistoric populations selected certain locations for single or repeated
occupation between the Early Archaic (8,500 BP) and Late Prehistoric (500 BP).
Explaining site selection is the driving research problem because location influences the
range of available economic opportunities and cultural adaptations. In order to address
this challenge a series of hypotheses have been formulated. These hypotheses are
designed around two general themes. The first theme is descriptive while the second
theme is interpretive.
Description oriented hypotheses focus on the descriptive character of the site
catchments to define the features related to site re-occupation and chronological
change. The first set of hypotheses is formulated to evaluate the physical
characteristics (topographic, physiographic, geological, great group and soil setting) of
158
site catchments in relation to occupation frequency (single, dual, tri, multi) and change
through time (from Early Archaic to Late Prehistoric II). This set of hypotheses is
formulated to provide a means to determine the specific physical characteristics
typically associated with sites that have been variably occupied and to provide a view of
how this changes through time. Evaluation of the relationships between site
catchments and the prehistoric activities that occurred at these locations require
hypotheses designed to examine site catchments in relation to site function (lithic
scatter, tool manufacture, hunting, food preparation) occupation density (ephemeral,
moderate, intensive) and cultural affinities (projectile point types). In order to factor in
the element of time, site function and occupation density are evaluated in relation to
occupation frequency and change through time.
Interpretive hypotheses are designed to test change through time and occupation
frequency in relation to potentialities for prehistoric utilization of certain environmental
resources. Environmental information is translated into a meaningful ecological context
related to human decision-making processes. Essentially, this portion of the analysis is
designed to test relationships between site locations and interpreted soil potentials.
Therefore the data are extended beyond the physical to the possible. Potential for
habitat suitable for certain types of wildlife and vegetation provides general insight on
what types of animals and vegetation were suited to certain soils.
Economic value extends interpretation beyond soils to vegetation linking these data
to possible prehistoric utilizations of plants that have the potential to exist given certain
soil types. Climatic interpretations link soils data to pollen data providing a means to
159
estimate the relative abundance of certain plants during wetter or drier climatic
intervals that have been documented in correlation with past patterns. Wildlife
(openland, rangeland, wetland); vegetation (grains and seed crops, grasses and
legumes, wild herbaceous plants); economic value (edibility, medicinal, supplemental);
and climatic interpretations of soils data are evaluated in relation to site catchment
characteristics, site contents, occupation frequency and change through time.
Relationships between the diversity within site catchment areas and the variability of
archaeological assemblages are examined. Variability between site catchments relative
to occupation frequency and change through time is evaluated. Defining the character
of the site catchment areas to determine diversity within and variability between these
areas helps explain why sites were selected repeatedly or ephemerally. Interpreting
the relative economic value of these variables provides a means to evaluate prehistoric
behavioral choices and adaptations as they change or stay the same through time. A
multi-scale investigation into landscape diversity within and variability between site
catchments of archaeological sites is selected to test these hypotheses. As such, the
SCA requires the ability to superimpose the archaeological record upon the micro-
environmental setting. Collectively, cultural and environmental data are analyzed, with
a unique methodology, to shed light on the character of multi-dimensional expression of
prehistoric decision-making processes.
As a part of the SCA process, environmental modeling is expected to reveal that
archaeological sites are strategically situated in areas that yield protection or in areas
that provide advantageous economic positioning. Correlations are expected between
160
select topographic settings, physiographic settings, soil potential, economic values and
climatic conditions. Given the physiographic character of north-central Texas, it is
expected that site selection strategies should reflect an equal utilization of the ecotones.
In relation to this, increased environmental diversity within the site catchment area is
expected. On the other hand, it is possible that certain areas were utilized for
specialized resource gathering. If this is the case, decreased environmental diversity is
expected. SCA also facilitates the inference of site function based on the economic
potential of the microenvironments that surround each site in conjunction with the
assemblages from them. It is anticipated that decisions made in relation to
environmental structure will provide comparable inferences regarding settlement
patterns to those drawn by other hunter-gatherer settlement studies.
In order to test the hypotheses archaeological and environmental data layers are
evaluated to test for differences between the environmental settings of site locations in
relation to assemblage composition. Occupation densities combined with activity
groups represent cultural behavior. Economic values (medicinal, edible, supplemental);
soil potential (vegetation and wildlife habitat); physiographic settings, topographic
settings, soil settings and geological settings are environmental attributes that are
factored into the statistical analyses. Hypotheses are either accepted or rejected on the
basis of how well the distribution of soils, geologic, hydrologic, topographic,
physiographic and economic variables met expectations in relation to assemblage
composition (lithic scatter, lithic manufacture, hunting, food preparation); occupation
density (ephemeral, moderate, intensive); occupation frequency (single, dual, tri, multi)
161
and change through time (Early Archaic-Late Prehistoric II). Variables and methods
utilized to evaluate each null hypothesis are addressed and summarized in this section.
Catchment diversity values, economic values, soil potentials, soil settings, great
groups, geological, physiographic and topographic data layers are utilized to test the
following hypotheses (table 5). The first predicts differences between site catchments
with respect to visitation frequency. The second evaluates change through time. The
third factors in site function. The fourth compares relationships and occupation density.
Null Hypothesis
There is no statistically significant difference between site catchments in relation to:
1 occupation frequency 2 change through time 3 site function 4 occupation density Table 5. Site catchment hypotheses.
Economic data layers are evaluated to look for relationships that may emerge regarding
specialized or generalized activities. Diversity data layers are evaluated to look for
relationships between catchment diversity and site occupation. Economic and diversity
values are both expected to correspond with increased site occupation frequency. An
ANOVA is conducted to evaluate economic values and diversity values in relation to
occupation frequency. A series of Chi-Square statistical tests are performed to test for
statistical significance of the great groups, soils, geological, topographic and
physiographic settings within site catchments in relation to occupation frequency. If
differences are statistically significant, frequency tables are evaluated to determine
what type of site selection patterns emerge in relation to these factors.
162
Chi-Square statistical tests are performed to test for statistical significance of site
function and occupation density in relation to change through time. If differences are
statistically significant, frequency tables are evaluated to determine what type of site
selection patterns are prevalent. These hypotheses are expected to shed light on the
relationship between site function and occupation density in relation to change through
time and occupation frequency (table 6) without factoring in the environment.
Null Hypothesis There is no statistically significant difference between site function
in relation to: 5 change through time 6 occupation frequency Null Hypothesis There is no statistically significant difference between occupation
density in relation to: 7 change through time 8 occupation frequency
Table 6. Site function hypotheses.
This facet is important because it provides a statistical view of prehistoric occupations
as they change through time. It reveals if inferred prehistoric activities and their
associated densities are similar or different with respect to change through time and in
relation to occupation frequency.
ANOVA is performed on a collection of databases that contain time diagnostic site
catchments and their associated proportional wildlife (openland, wetland, and
rangeland) potentials; vegetation (grains and seed crops, grasses and legumes, wild
herbaceous plants) potentials; and economic values (edible, medicinal, supplemental).
This multi-variate statistical test will evaluate a series of hypotheses (Table 7).
163
Null Hypothesis There is no statistically significant difference between wildlife potentials associated with site catchments in relation to:
9 change through time 10 occupation frequency Null Hypothesis There is no statistically significant difference between
vegetation potentials associated with site catchments in relation to:
7 change through time 8 occupation frequency Null Hypothesis There is no statistically significant difference between economic
values associated with site catchments in relation to: 9 change through time 10 occupation frequency
Table 7. Economic values hypotheses.
Determining the relationship between the character of site catchment areas and the
variability of archaeological assemblages relative to occupation frequency and change
through time provides explanatory mechanisms for prehistoric settlement patterns.
Purposeful behavior is patterned and based on socially transmitted information that
facilitates successful adaptations. If site selection is the result of purposeful human
activity, it should likewise be patterned. Although it is expected that the patterning of
site selection is reflective of the human behavior that produced it, these similarities
must be demonstrated and not assumed. Human actions are processes in time and
space that have structure and can be presented as a network of inter-related behavior.
Inferences can be made based on what is known about the character of the
environments that provided the range of possibilities for prehistoric adaptations. When
site distribution is calibrated with the environmental setting, given a variety of
scenarios, it is possible to build a model of regional prehistoric settlement patterns.
164
Study Area
The area in north-central Texas (figure 30), selected for this research, is defined by
the three major tributaries of the Trinity that intersect three distinctive physiographic
zones. For the most part sites are situated along or near these waterways. On a
regional scale, SCA is performed on 150 surface sites selected from the West Fork, Elm
Fork and East Fork of the Trinity, with projectile point typology as the means to
establish chronology and occupation frequency. Data gathered from these regional
site locations are evaluated using 10 excavated sites from the Elm Fork of the Trinity
that produce a well-documented and well-stratified chronological record. These
excavated sites provide a means of temporal control for the regional model. The time
span of this study ranges from 8500 years before present, during the Early Archaic, to
the Neo-American era, 500 years before the present era. Excavated sites selected for
this research provide a discontinuous record of occupations, in north-central Texas,
from the Early Archaic to Late Prehistoric. Excavated sites are useful for this research
since they provide a detailed account of artifacts, features, fauna and radiocarbon
dated materials. These materials are particularly suited to SCA because the level of
detail facilitates a more complete reconstruction of economic activities in space and
through time. In addition, excavated sites provide empirical field data as a means to
test validity of the extrapolated data derived from digital map layers that form the basis
for the analysis. Selected study sites facilitate a chronological examination of diversity
within site catchment areas and variability between site locales to consider the variety
of factors that could influence site selection. In addition, an opportunity to examine site
165
Figure 30. North-central Texas research study area.
166
catchments for evidence of functional constants that may be associated with site
selection is provided. The regional scale analysis that includes the 150 surface sites
provides a basis from which to infer site type, function, examine trends and evaluate
the environmental diversity and relative economic value on a regional scale. According
to Ferring (1986), archaeologists must continue to develop research methods that
enhance the study of intra-site variability with respect to the larger, multi-site cultural
system. Although, Ferring is referring to artifacts and fauna, the same concept is true
for the environmental setting. This research views the catchment area as an extension
of the archaeological remains. As such the results of the SCA are an important facet of
the regional settlement pattern model formulated for this region.
167
CHAPTER 6
RESULTS
The most analytically informative data produced by conducting the site catchment
analysis (SCA) in north-central Texas is contained by a series of maps, graphs and
tabular data. In this chapter results are evaluated with respect to each variable that is
evaluated either spatially or statistically and contributes to the development of the
regional settlement pattern model. A discussion of the distribution of sites in relation to
the geological, physiographic, topographic and soils settings, as well as, site function
and site activities are followed by a synopsis of the statistical results.
Geological Setting
The geological setting of sites in the study area (figure 31) provides tremendous
utility for environmental reconstruction and behavioral interpretation. Closer observation
of this variable reveals strategic decision-making. Sites located in the West Fork study
area are predominantly situated in sandstone settings with secondary presence in
alluvial settings and tertiary presence in limestone (figure 32-33). Elm Fork sites follow
a similar pattern however shale settings replace limestone settings. East Fork sites are
typically associated with alluvial settings, marl or a combination therein.
In the Elm Fork study area 48% of the Archaic sites are on fluvatile or alluvial
deposits, 41% on sandstone and 9% on shale. Late Prehistoric sites are situated in
water deposited sediments (40%), sandstone (45%) and shale (13%). Hunting sites in
the Elm Fork area are fairly equally represented in both water deposited sediments
(44%) and sandstone settings (37%) with the remaining in shale (9%) and limestone
168
(3%) settings. Lithic manufacture camps are primarily associated with sandstone
(45%) with a secondary presence (34%) in alluvial and fluvatile terrace deposits.
In the West Fork area 56% of the Archaic sites are associated with sandstone and
22% with water deposited sediments. Of the Late Prehistoric sites 25% are found in
association with alluvium or fluvatile terrace deposits and 50% with sandstone. Sites
that contain evidence of lithic manufacture, hunting and even base camps are found in
association with sandstone (75%) with the remaining 25% found on fluvatile terrace
deposits. Only one site was found in relation to limestone and upon closer inspection it
appears to be situated directly on the dividing line between the Goodland Limestone
Formation and the Antlers Sandstone. Sites in the West Fork area are actually a very
good example of sites oriented in an ecotonal setting with a notable precision.
Interestingly, in almost every case cases the sites are on the sandstone side of the
ecotone about a half of a kilometer from the limestone. The alternative scenario is
illustrated by one site in particular situated almost half way between the West Fork and
the Elm Fork. In this case, where Goodland limestone dominates the setting, a small
fluvatile terrace deposit was selected and the site was placed on the edge overlooking a
stream valley. This is an excellent example of how geology can be used to reconstruct
prehistoric behaviors that appear to be influenced by past environments.
In the East Fork portion of the study area Archaic sites tend to be associated with
alluvium (27%) and fluvatile terrace deposits (33%). An opposite pattern occurs during
the Late Prehistoric when sites are primarily located in alluvial deposits (34%) with a
169
Figure 31. Geological Setting of Sites in Study Area.
170
Figure 32. Geological Settings and Time Periods. This figure illustrates the relationships between the number of sites that contain each particular geological deposit. For the sake of clarity in illustration combined with the fact that there is a significant overlap of site selection with respect to change through time it was necessary to consolidate the subdivisions of the particular time periods. Early, Middle and Late Archaic are grouped into “Archaic”. The first and second phase of the Late Prehistoric is combined into the “Late Prehistoric” category. The “Prehistoric” “category consists of sites that do not sites are associated with marl. Although functional differences between the Prehistoric
GEOLOGICAL SETTING: ELM FORK
10
20
30
40
50
60
WATERDEPOSITEDSEDIMENTS
LIMESTONESETTINGS
SHALESETTINGS
CLAY RICHLIMESTONESETTINGS
SANDSTONESETTINGS
SIT
ES (
CO
UN
TS)
ARCHAIC LATE PREHISTORIC PREHISTORIC
GEOLOGICAL SETTING: WEST FORK
10121416182022
WATERDEPOSITEDSEDIMENTS
LIMESTONESETTINGS
SHALESETTINGS
CLAY RICHLIMESTONESETTINGS
SANDSTONESETTINGS
SIT
ES
(CO
UN
TS)
ARCHAIC LATE PREHISTORIC PREHISTORIC
contain diagnostic materials. Geological deposits are also consolidated into their more generalized “Parent” groups.
171
Frc
GEOLOGICAL SETTING: ELM FORK
10
20
30
40
50
60
WATERDEPOSITEDSEDIMENTS
LIMESTONESETTINGS
SHALESETTINGS
CLAY RICHLIMESTONESETTINGS
SANDSTONESETTINGS
SIT
ES (
COU
NTS
)
MULTI-COMPONENT SINGLE-COMPONENT
GEOLOGICAL SETTING: WEST FORK
10121416182022
WATERDEPOSITEDSEDIMENTS
LIMESTONESETTINGS
SHALESETTINGS
CLAY RICHLIMESTONESETTINGS
SANDSTONESETTINGS
SIT
ES (
COU
NTS
)
MULTI-COMPONENT SINGLE-COMPONENT
igure 33. Geological Settings and Occupation Frequency. This figure illustrates the elationship between geological deposits and occupation frequency. Sites are onsolidated into single or multiple occupancy and geology into generalized lithology.
172
secondary presence in fluvatile terrace deposits (19%). In addition, 26% of the Late
Prehistoric sites are associated with marl. Although functional differences between the
sites along the East Fork are not currently clear due to lack of data there is a strong
correlation between the four sites that contain the "Wylie Focus Pits" and the Ozan Marl
formation. This is significant due to the fact that most of the landscape is dominated
by the Austin Chalk to the West and Marlbrook Marl to the East. However, given that
alluvial deposits and the associated major tributaries are carved into the Ozan Marl, this
appears to be a logical choice that provided immediate access to water resources. In
addition, this geological deposit is less resistant than the surrounding geology which
would have facilitated the construction of the characteristic deep ceremonial pits.
Geology in correlation with occupation frequency follows similar patterns as change
through time. This is probably due to the fact that sites that exhibit reoccupation are
occupied during more than one time period and the majority of sites with diagnostic
materials exhibited reoccupation. This is a problematic facet of the current sample. As
such, it is emphasized that the bias in site distribution is influencing the overall trends
but when the sites are observed individually there is clearly a pattern of strategic site
selection with respect to specific geological settings. In this case, sites are typically
situated in water deposited sediments nested within sandstone settings.
Physiographic Setting Physiographic (figure 34) settings of the Elm Fork include three vegetation zones.
East to West they are: Blackland Prairie, Eastern Cross Timbers and Grand Prairie or
Fort Worth Prairie. Most of the Elm Fork of the Trinity is included in the Eastern Cross
173
Timbers. The Elm Fork of the Trinity includes a wide band of the Eastern Cross Timbers
flanked on the west by the Grand Prairie and on the east by the Blackland Prairie.
Regardless of the effects of change through time (figure 35) and occupation frequency
(figure 36) most of the archaeology along the Elm Fork is in the ecotone between the
Cross Timbers and Prairies.
The West Fork of the Trinity (figure 34) is in a transitional zone between the
Western Cross Timbers and the Fort Worth/Grand Prairie. The cultures that occupied
this area lived precisely within the eco-tone between woodlands and prairies even when
change through time (figure 35) and occupation frequency (figure 36) are considered.
Sites situated in these areas have a high degree of flexibility between the prairies and
the woods increasing the success of various subsistence strategies.
The East Fork of the Trinity is situated in the center of the Blackland Prairie. The
only access to woodland species is along riparian corridors. As such sites tend to
aggregate along the main stem of the East Fork (figure 34) rather than along minor
tributaries and ephemeral streams. Bottomland hardwoods also provide access to the
benefits offered by a wooded environment. Physiographically, the options open to
prehistoric populations were limited. However, the variation in the limestone and marl
geology, combined with the riparian settings, provided a fairly diverse environment.
Utilization of the ecotone occurs along both the West Fork and the Elm Fork.
Change through time appears to be more affected by the number of known sites than
due to prehistoric preferences. Ecotonal settings on the fringes of the eastern and
western edges of the Cross Timbers appear to be the most heavily utilized areas along
174
the Elm Fork. Sites are equally distributed between the two prairie/woodland ecotones
with a scatter of centrally located sites in the Cross Timbers.
Figure 34. Physiographic settings in north-central Texas study area.
175
Minor riparian corridors (41%) in both the West and the Elm Forks are utilized most
often. Major River corridors (30%) and ephemeral stream settings (29%) are utilized
fairly equally. Along the East Fork there are no ecotonal areas immediately present
unless the riparian environments are included in this category. East Fork sites are
PHYSIOGRAPHIC SETTING: ELM FORK
1020304050607080
PRAIRIES CROSSTIMBERS
MAJORRIPARIAN
CORRIDORS
MINORRIPARIAN
CORRIDORS
ECOTONE
SITE
(CO
UN
TS)
ARCHAIC LATE PREHISTORIC PREHISTORIC
PHYSIOGRAPHIC SETTING: WEST FORK
10121416
18202224
PRAIRIES CROSSTIMBERS
MAJORRIPARIAN
CORRIDORS
MINORRIPARIAN
CORRIDORS
ECOTONE
SITE
(CO
UN
TS)
ARCHAIC LATE PREHISTORIC PREHISTORIC
Figure 35. Physiographic settings and time periods.
176
PHYSIOGRAPHIC SETTING: ELM FORK
1020
30405060
7080
PRAIRIES CROSSTIMBERS
MAJORRIPARIAN
CORRIDORS
MINORRIPARIAN
CORRIDORS
ECOTONE
SIT
E (
CO
UN
TS)
MULTI-COMPONENT SINGLE-COMPONENT
PHYSIOGRAPHIC SETTING: WEST FORK
1012
1416
1820
2224
PRAIRIES CROSSTIMBERS
MAJORRIPARIAN
CORRIDORS
MINORRIPARIAN
CORRIDORS
ECOTONE
SIT
E (C
OU
NT
S)
MULTI-COMPONENT SINGLE-COMPONENT
Figure 36. Physiographic settings and occupation frequency.
situated primarily within grasslands along major tributaries (26%), minor tributaries
(26%) and ephemeral streams (52%) providing immediate access riverine resources.
Observations from previous investigations that reported grassland, woodland,
riverine and ecotonal utilization are substantiated. The observation that sites are
situated to equally utilize forest, riparian and prairie environments appears to be true in
177
relation to the Elm and West Forks but is not the case along the East Fork. Although
there is inherent ecotonal utilization in the East Fork area by virtue of the riparian
corridors nested within the prairie environment. Given the spatial patterning and
reoccupation of sites through time it is likely that the environment was a more
significant factor then socially defined boundaries. Although, it is also possible, very
much like today, that environmental boundaries coincided with social boundaries. It
appears that these populations strategically placed themselves so they would have
access to a variety of economically valuable resources.
Topographic Setting
Topographic settings of site catchments in the Elm Fork (figure 39) and West Fork
(figure 40) provide a means to compare the orientation of sites from a variety of
perspectives. Archaic and Late Prehistoric sites located on the West Fork are
predominantly situated in terrace (47%) and upland (26%) settings near primary and
secondary water sources. Drainage settings (21%) and floodplain (5%) settings are
represented by 3 single and 2 multi-component sites. Archaic and Late Prehistoric Elm
Fork sites are predominantly located in upland (45%) and terrace (41%) settings. Only
12% are located in the floodplain and 1% in drainages. Archaic East Fork populations
selected alluvial terraces (41%), floodplains (25%), and uplands (33%). During the
Late Prehistoric a slight decrease in terrace settings to 38% is countered by an increase
in floodplains to 30% while upland settings (32%) remain about the same. Fairly
equal distribution of sites in upland and terrace settings in both the West and the Elm
Fork exist regardless of change through time (figure 37). Overlapping of time
178
diagnostic sites creates complexity in this respect. A similar distribution between
topographic settings occurs in relation to occupation frequency that is fairly consistent
between the East, West and Elm Forks of the Trinity (figure 38).
TOPOGRAPHIC SETTING: ELM FORK
10
20
30
40
50
60
70
BOTTOMLANDS WATER(DRAINAGES)
FLOODPLAINS UPLANDS TERRACES
SIT
E (
COU
NT
S)
ARCHAIC LATE PREHISTORIC PREHISTORIC
TOPOGRAPHIC SETTING:WEST FORK
1012141618202224
BOTTOMLANDS WATER(DRAINAGES)
FLOODPLAINS UPLANDS TERRACES
SITE
(C
OU
NT
S)
ARCHAIC LATE PREHISTORIC PREHISTORIC
Figure 37. Topographic Settings and Time Periods.
179
TOPOGRAPHIC SETTING: ELM FORK
1020304050607080
BOTTOMLANDS WATER(DRAINAGES)
FLOODPLAINS UPLANDS TERRACES
SIT
ES
(C
OU
NTS
)
MULTI-COMPONENT SINGLE-COMPONENT
TOPOGRAPHIC SETTING: WEST FORK
1012141618202224
BOTTOMLANDS WATER(DRAINAGES)
FLOODPLAINS UPLANDS TERRACES
SIT
ES
(CO
UN
TS
)
MULTI-COMPONENT SINGLE-COMPONENT
TOPOGRAPHIC SETTING (EAST FORK)
05
1015202530354045
ALLUVIAL TERRACES FLOODPLAINS UPLANDS STREAM TERRACES
SIT
ES
(C
OU
NTS
)
MULTI-COMPONENT SINGLE-COMPONENT
Figure 38. Topographic Settings and Occupation Frequency.
180
Figure 39. Elm fork site catchment elevations, change through time and projectile frequency. This figure is a screen shot of the GIS interface where the sites are evaluated from two perspectives (change through time and frequency of diagnostic points) juxtaposed over top the elevation. The larger circles are meant to show the clusters that are occurring per time period. With a few exceptions, catchments outside of the clusters are non-diagnostic general prehistoric sites.
181
Figure 40. West fork site catchment elevations, change through time and projectile frequency. This figure is a screen shot of the GIS interface where the sites are evaluated from two perspectives (change through time and frequency of diagnostic points) juxtaposed over top the elevation. In the West Fork portion of the study area there is very little diagnostic evidence as reflected by the lack of indication of change through time. Only two sites contain evidence of change through time.
182
Results support the observations made during previous investigations that suggest that
prehistoric populations primarily selected sites located in terraces and upland settings in
ecotonal transitional zones between prairie and woodland settings. Terrace and upland
settings along both the West and Elm Fork of the Trinity indicate reoccupation. East
Fork sites contain evidence of occupation in uplands, floodplains, and terrace settings
near bottomland riparian environments. In this respect the East Fork is very similar to
the West and Elm Forks of the Trinity.
Soils: Great Groups Soil taxonomy provides a means to interpret the potential of soils to produce
vegetation and provide habitat for wildlife. Given the similar geological foundations of
the Elm and West Fork results indicate a fairly equal correlation with Alfisols, Mollisols
and Vertisols. Paleustalfs, Haplusterts, Haplustolls, Haplustalfs, Hapluderts and
Calciustolls. All of these great groups have a common denominator that indicates a
fairly moist environment with dry intervals probably related to seasonality or climatic
fluctuations. Evaluating soil distribution (figure 41) for the West and Elm Forks across
the study area and in the diagnostic catchment areas provides an overview of the
variability. It is the distribution of these soils in relation to sites that determines the
relative accessibility to economically valued resources by prehistoric populations. The
primary difference between the great groups in the study area regardless of change
through time (figure 43) and occupation frequency (figure 44) is in the level of horizon
development and soil order. Otherwise in all portions of the study area there is a
proportional distribution of Alfisols, Mollisols, Vertisols and Inceptisols.
183
Figure 41. Great group distribution and site catchments.
184
East Fork soils (figure 42) are considerably different from the West Fork and Elm Fork.
Pellusterts, Chromuderts and Chromusterts indicate a strong presence of Vertisols with
varying degrees of organic content. The ustic component indicates a fairly moist
setting with dry intervals whereas the udic setting indicates a higher moisture regime
Figure 42. East Fork great group distribution and site catchments.
185
GREAT GROUPS: ELM FORK
1020304050607080
MOLLISOLS VERTISOLS ALFISOLS ENTISOLS ANDINCEPTISOLS
SIT
ES (
COU
NT
S)
ARCHAIC LATE PREHISTORIC PREHISTORIC
GREAT GROUPS: WEST FORK
10
15
20
25
MOLLISOLS VERTISOLS ALFISOLS ENTISOLS ANDINCEPTISOLS
ARCHAIC LATE PREHISTORIC PREHISTORIC
GREAT GROUPS: EAST FORK
-5
5
15
25
35
45
VERTISOLS MOLLISOLS ALFISOLS INCEPTISOLS
SIT
ES (
COU
NT
)
ARCHAIC LATE PREHISTORIC PREHISTORIC
Figure 43. Great groups and time periods.
186
year-round. Haplaquolls and Haplustolls represent the typical Mollisols that are
associated with the Austin Chalk. Hapluquolls are associated with a wetter environment
then the Haplustolls probably indicating a closer proximity to the proximal water
resources. Higher organic content is associated with these taxonomic units which is
attributable to vegetation associated with the characteristic prairie setting.
Comparatively East Fork sites provide a significant contrast to the West and Elm Fork
of the Trinity. The East Fork is dominated by Vertisols with a secondary presence of
Mollisols and tertiary presence of Alfisols regardless of change through time (figure 43).
In relation to this it is interesting to note that multi-component sites tend to have a
fairly equal distribution of great groups indicating that these populations were probably
strategically selecting base camp locations in areas that provided access to the highest
diversity of vegetal resources. On the other hand single component (figure 44) site
catchments all contain Vertisols. West Fork single component sites show more variation
in the great groups within their respective catchments whereas the multi-component
sites have more of an equal distribution of great group types with a slightly greater
emphasis of Alfisols. Given that these sites are associated with sandstone settings this
situation is anticipated. In addition, the number of sites situated on the ecotone, along
the Elm Fork, there is a fairly equal distribution of great groups within the site
catchments regardless of occupation frequency (figure 44). Soils provide a situation that
is quite different from the more generalized categories. This variable has much more
overlap than any other due to the level of detail it provides. In all cases site
catchments contain more than one soil series and more than one great group.
187
GREAT GROUPS: ELM FORK
1020304050607080
MOLLISOLS VERTISOLS ALFISOLS ENTISOLS ANDINCEPTISOLS
SITE
S (C
OU
NTS
)
MULTI-COMPONENT SINGLE-COMPONENT
GREAT GROUPS: WEST FORK
0
2
46
8
10
12
MOLLISOLS VERTISOLS ALFISOLS ENTISOLS ANDINCEPTISOLS
SIT
ES (
COU
NT
S)
MULTI-COMPONENT SINGLE-COMPONENT
GREAT GROUPS: EAST FORK
0
10
20
30
40
VERTISOLS MOLLISOLS ALFISOLS INCEPTISOLS
SIT
ES (
CO
UN
TS)
MULTI-COMPONENT SINGLE-COMPONENT
Figure 44 Great groups and occupation frequency.
188
Soils: Soil Settings
Soil settings in the Elm and West Fork are fairly similar. Due to the level of detail
provided by the soil surveys the East Fork is not included in this portion of the analysis.
The most dominant groups are clay-rich and loamy settings. Sandy soils for the most
part have a high clay content which has created soils that are loamy in character. There
appears to be an equal representation of each setting along the Elm and West Forks in
relation to change through time (figure 45) and occupation frequency (figure 46).
SOIL SETTING: ELM FORK
1020304050607080
CLAY RICHENVIRONMENTS
SANDYENVIRONMENTS
LOAMYENVIRONMENTS
OTHER
SITE
S (C
OU
NTS
)
ARCHAIC LATE PREHISTORIC PREHISTORIC
SOIL SETTING: WEST FORK
1012141618202224
CLAY RICHENVIRONMENTS
SANDYENVIRONMENTS
LOAMYENVIRONMENTS
OTHER
SITE
S (C
OU
NTS
)
ARCHAIC LATE PREHISTORIC PREHISTORIC
Figure 45. Soil settings and time periods.
189
SOIL SETTING: ELM FORK
10
20
30
40
50
60
70
80
CLAY RICHENVIRONMENTS
SANDYENVIRONMENTS
LOAMYENVIRONMENTS
OTHER
SIT
ES (
COU
NT
S)
MULTI-COMPONENT SINGLE-COMPONENT
SOIL SETTING: WEST FORK
10
12
14
16
18
20
22
24
CLAY RICHENVIRONMENTS
SANDYENVIRONMENTS
LOAMYENVIRONMENTS
OTHER
SIT
ES
(CO
UN
TS
)
MULTI-COMPONENT SINGLE-COMPONENT
Figure 46 Soil settings and occupation frequency. Soil settings from the West and Elm Forks reflect the utilization of eco-tonal settings.
These characteristic settings are essentially the basis for vegetation correlations. As
such, the frequency of sites associated with a given soil setting provides important
information regarding prehistoric preferences of habitat. The general categories
illustrated here are intended to show overall trends. Interpretive views are available
when this data is evaluated in correlation with the types of associated economic values.
190
Soils: Economic Interpretations
Economic interpretations for soils provide enlightening information in relation to site
distribution. While there is not a statistically significant difference, between site
catchments, there are similarities that indicate strategic selection of locations on the
landscape that provide important resources to sustain prehistoric populations. One of
the few statistically significant results between site catchments are associated with soil
settings. This is directly related to the particular economic value of the site catchment
as established by the current research. Mapping of the distributions of economic value
with respect to edibility, medicinal value, supplemental value, wildlife and vegetation
provides visual cues that reveal patterns. These patterns can be viewed in relation to
change through time, occupation frequency, density, projectile point distribution and a
variety of other facets. A series of maps are provided in the appendix (A1-A22) to
illustrate the relationships of the sites to the resources relative to their interpretive
values as site location changes through time. Maps provided in this section illustrate
relationships between soil values and occupation frequency, as well as, site locations
and varying degrees of economically valuable resources. Graphs provided in this section
summarize data extracted from the site catchment areas in relation to occupation
frequency and change through time.
Sites are evaluated in relation to the relative economic values represented within the
catchment areas. Soils data are recoded to illustrate (figure 47) the overall distribution
of edible, medicinal and supplemental vegetation patterns on a larger scale in relation
to the site locations. Darker values indicate a lower potential for these specific types of
191
Figure 47. Economic values and prehistoric components.
192
vegetation whereas the lighter values indicate an abundance of the associated
resources. The lower values are weighting the sample due to the lack of soils data
where reservoirs now exist. Change through time with respect to these variables is
illustrated in the appendix. Overall, there is a fairly similar distribution of each group
within the site catchment areas although the abundance varies.
Sites that contain evidence of repeated occupation (figure 48) indicate a strong
presence of edible vegetation followed by supplemental and medicinal along the West
Fork and the Elm Fork. However, the Elm Fork values reveal a relatively higher edibility
factor. With respect to change through time (figure 49) similar patterns occur. During
the Middle Archaic and Late Prehistoric I medicinal values rise above supplemental
values along the Elm Fork. The East Fork situation is similar except during the Late
Prehistoric I era medicinal values notably jump above the supplemental values reaching
almost the same level as the edibility values and then gradually decrease. It is likely
that these locations provided seasonal resources that were complimented by a variety
of edible and medicinal plants paired with woodland resources as reflected by
supplemental values. The supplemental value category primarily reflects wood
utilization. As such, sites situated in the Cross Timbers are significantly weighted with
respect to the supplemental category. The same situation is true for sites located along
riparian corridors where woody vegetation tends to aggregate. Fluctuations through
time indicate some type of pattern in relation to the correlation between particular
economic values and site locations. This could be attributed to climate change,
territoriality, seasonality or simply cultural preferences.
193
West Fork Economic Values and Occupation Frequency
0.0350
0.0400
0.0450
0.0500
0.0550
0.0600
SINGLE DUAL TRI MULTI
Edibility Medicinal Supplemental
Elm Fork Economic Values and Occupation Frequency
0.0350
0.0400
0.0450
0.0500
0.0550
0.0600
SINGLE DUAL TRI MULTI
Edibility Medicinal Supplemental
Figure 48. Economic values and occupation frequency.
194
West Fork Economic Values and Change Through Time
0.0350
0.0400
0.0450
0.0500
0.0550
0.0600
ARCH
AIC
EARL
Y AR
CHAI
C
MIDDLE
ARC
HAIC
LATE
ARC
HAIC
LATE
PREH
ISTO
RIC
LATE
PREH
ISTO
RIC
I
LATE
PREH
ISTO
RIC
II
PREH
ISTO
RIC
Edibility Medicinal Supplemental
Elm Fork Economic Values and Change Through Time
0.0350
0.0400
0.0450
0.0500
0.0550
0.0600
ARCH
AIC
EARL
Y AR
CHAI
C
MIDDLE
ARC
HAIC
LATE
ARC
HAIC
LATE
PREH
ISTO
RIC
LATE
PREH
ISTO
RIC
I
LATE
PREH
ISTO
RIC
II
PREH
ISTO
RIC
Edibility Medicinal Supplemental
Figure 49. Economic values and time periods.
195
Mapped distributions of potentials for vegetation types given certain soil combinations
indicate relationships between site locations and concentrations of grains and seed
crops, grasses and legumes, and wild herbaceous plants (figure 50). This figure
illustrates the distribution of soils that have been recoded to represent wild herbaceous
plants, grains and seed crops and medicinal grasses/legumes. The darker values
indicate a lower potential for these types of vegetation whereas the lighter values
indicate an abundance of these resources. Lower values are weighting the sample due
to the lack of soils data where reservoirs now exist. Change through time in relation to
these variables is illustrated in the appendix.
All of the sites in the study area appear to have a stronger relationship with wild
herbaceous plants than any other vegetation group. These types of plants are primarily
associated with forested settings. This relationship is not surprising given that
prehistoric populations appear to have aggregated in the Cross Timbers and in riparian
environments. Grains and seed crops are secondary in relation to change through time
yet tertiary in relation to occupation frequency. This is an interesting relationship that
may indicate that diagnostic sites are found more often in association with this type of
vegetation whereas single occupancy sites (mostly non-diagnostic) are situated in
grassier areas. West Fork sites have a strong correlation with Grasses and Legumes
given that there are sites situated on the ecotone near prairie settings. This type of
vegetation is in the proximity of the Elm Fork sites although it is not concentrated in the
site catchment areas. Overall, relationships between the vegetation types remain fairly
similar with respect to occupation frequency (figure 51) and time period (figure 52).
196
Figure 50. Vegetation potential and prehistoric components.
197
West Fork Vegetation Potential and Occupation Frequency
0.035
0.085
0.135
0.185
0.235
0.285
0.335
0.385
0.435
0.485
SINGLE DUAL TRI MULTI
Grasses and Legumes Grains and Seed Crops Wild Herbaceous Plants
Elm Fork Vegetation Potential and Occupation Frequency
0.035
0.085
0.135
0.185
0.235
0.285
0.335
0.385
SINGLE DUAL TRI MULTI
Grasses and Legumes Grains and Seed Crops Wild Herbaceous Plants
Figure 51. Vegetation potential values and occupation frequency.
198
West Fork Vegetation Potential and Change Through Time
0.0350.0850.1350.1850.2350.2850.3350.3850.4350.485
ARCH
AIC
EARL
Y AR
CHAI
C
MIDDLE
ARC
HAIC
LATE
ARC
HAIC
LATE
PREH
ISTO
RIC
LATE
PREH
ISTO
RIC
I
LATE
PREH
ISTO
RIC
II
PREH
ISTO
RIC
Grasses and Legumes Grains and Seed Crops Wild Herbaceous Plants
Elm Fork Vegetation Potental and Change Through Time
0.0350.0850.1350.1850.2350.2850.3350.3850.435
ARCH
AIC
EARL
Y AR
CHAI
C
MIDDLE
ARC
HAIC
LATE
ARC
HAIC
LATE
PREH
ISTO
RIC
LATE
PREH
ISTO
RIC
I
LATE
PREH
ISTO
RIC
II
PREH
ISTO
RIC
Grasses and Legumes Grains and Seed Crops Wild Herbaceous Plants
Figure 52. Vegetation potential values and time periods.
199
Soil potential to provide habitat for certain types of wildlife as estimated by soil survey
data provides a view of another facet of potentiality. Mapped values reveal differential
distribution that indicates that certain types of wildlife are more likely to be
concentrated in particular locations (figure 53). This figure illustrates the distribution of
soils that have been recoded to represent relative potential for rangeland, openland and
wetland wildlife habitat. The darker values indicate a lower potential for these types of
vegetation whereas the lighter values indicate an abundance of these resources. Lower
values are weighting the sample due to the lack of soils data where reservoirs now
exist. Change through time in relation to potential for wildlife habitats is illustrated in
the appendix.
Sites that contain evidence of variable occupation frequency follow similar trends
(figure 54). This is probably due to sample size combined with the fact that sites are
often clustered along the Elm Fork and have overlapping site catchment areas which
influence the outcome. Through time there are fairly consistent correlations between
site location and openland, wetland and rangeland wildlife habitat (figure 55). Wetland
habitat remains fairly low due to the low concentrations of soils associated with these
settings. This is not to say that wetland wildlife was not utilized as much as the other
types. In this case soil classification is weighting the sample. Riverine corridors are not
factored into the distribution of wetland wildlife. Based on the archaeological faunal
data it is known that prehistoric populations, in this region, utilized aquatic resources,
regardless of the time period. Given the limited sample it is difficult to know if this is
the case with all the sites but the proximity to water resources favorably supports this.
200
Figure 53. Wildlife potential and prehistoric components.
201
Elm Fork Wildlife Potental and Occupation Frequency
0.001
0.051
0.101
0.151
0.201
0.251
0.301
0.351
0.401
SINGLE DUAL TRI MULTI
Openland Wildlife Rangeland Wildlife Wetland Wildlife
West Fork Wildlife Potential and Occupation Frequency
0.001
0.051
0.101
0.151
0.201
0.251
0.301
0.351
0.401
SINGLE DUAL TRI MULTI
Openland Wildlife Rangeland Wildlife Wetland Wildlife
Figure 54. Wildlife potential values and occupation frequency.
202
West Fork Wildlife Potential and Change Through Time
0.0010.0510.1010.1510.2010.2510.3010.3510.401
ARCH
AIC
EARL
Y AR
CHAI
C
MIDDLE
ARC
HAIC
LATE
ARC
HAIC
LATE
PREH
ISTO
RIC
LATE
PREH
ISTO
RIC
I
LATE
PREH
ISTO
RIC
II
PREH
ISTO
RIC
Openland Wildlife Rangeland Wildlife Wetland Wildlife
Elm Fork Wildlife Potental and Change Through Time
0.0010.0510.1010.1510.2010.2510.3010.3510.401
ARCH
AIC
EARL
Y AR
CHAI
C
MIDDLE
ARC
HAIC
LATE
ARC
HAIC
LATE
PREH
ISTO
RIC
LATE
PREH
ISTO
RIC
I
LATE
PREH
ISTO
RIC
II
PREH
ISTO
RIC
Openland Wildlife Rangeland Wildlife Wetland Wildlife
Figure 55. Wildlife potential values and time periods.
203
Site Catchment Diversity Sites catchments are measured for relative degrees of diversity based on the
proportional presence of soil series contained in the 1km area. Results (figure 39)
indicate the largest amount of variability in sites that contain evidence of single
occupation. Given the number of sites in this category it is not surprising and in fact
anticipated. However, there are considerably less multi-component sites and this
group also exhibits increased variability.
Figure
parti
al fo
r VIS
ITA
TIO
N
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
VISITATION
Dual
Multi
Single
Tri
56. ANOVA comparisons of site catchment diversity and occupation frequency.
204
With respect to change through time the most significant range of variation occurs
with the Late Archaic sites. The results illustrated by the graph (figure 40) indicate that
there are expressed changes that occur in relation to site catchment diversity. It is
possible that during the Late Archaic populations diversified and enjoyed a greater
range of motion. Previous studies have indicated in prior research that patterns of site
location indicate a decrease in territorial boundaries during the Late Prehistoric. If
these populations were restricted to certain locations under variable climatic conditions
there may be relatively less diversity. However, given the character of the region a fair
amount of diversity is available within the woodlands, prairies, ecotones and riverine
settings such that a certain degree will be continuously present.
parti
al fo
r AG
E
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
AGE
A EA LA LP LPI
LPII
MA P
Figure 57. ANOVA comparisons: Site catchment diversity and time periods.
205
Climatic Interpretations
Climatic conditions are often referred to as explanatory mechanisms for culture
change. Recoded soils data based on pollen records reflects relative vegetal abundance
given various climatic conditions (figure 55). When these data are mapped correlations
Figure 58a. Distribution of climatically abundant vegetation families.
206
can be drawn in relation to site distribution as it changes through time (figure 55 and
figure 56). The working assumption is that populations would have been drawn to
locations that contain abundant opportunities of valuable resources given the
appropriate climatic conditions. This is a generalized model that could be calibrated
with more detailed data. However, due to time constraints it is generalized for the
purpose of illustration. If these patterns are accurate then it is suggested that sites
that were selected by prehistoric populations would have provided a fairly stable
economic resource base in spite of climatic fluctuations. Even though changes through
time are apparent site locations are consistently situated near abundant resources.
Figure 58b. Distribution of climatically abundant vegetation families.
207
Site Activities and Density
Prehistoric activities are inferred based on the types of tools recovered from given
site locations. When they are viewed in relation to the various characteristics of the site
catchments the dynamic inter-relationships between prehistoric activities and the
environment within which they were performed can be evaluated. Sites that contain
evidence of tool manufacturing activities, food preparation, hunting activities and lithic
scatter are predominantly represented in fluvatile and alluvial deposits. In some cases
sites contain evidence of all four activities and as such would be considered to be base
camps. However, the degree of activity that is occurring in a given area can be cross-
checked with the level of occupation density. Correlations indicate at least twelve site
locations exhibit intensive occupation in upland and terrace settings that are associated
with fluvatile and alluvial deposits. Ephemeral sites are primarily correlated with lithic
scatter which are situated in terrace and upland settings situated on fluvatile, alluvial
and sandstone deposits along riparian corridors in the Cross Timbers. Activities
represented at West Fork sites indicate an increased frequency of hunting during the
Archaic and equal distribution of tool manufacture, hunting and food preparation during
the Late Prehistoric. Given that prehistoric sites are non-diagnostic sites defined by
lithic scatter this activity is pronounced for both the West and Elm Forks during the
general Prehistoric. Along the Elm Fork hunting and food preparation are predominant
during the Archaic with a similar pattern accompanied by a slight increase in food
preparation occurring during the Late Prehistoric. Activities in relation to the variables
addressed in this section are summarized by a series of graphs (Figures 59-65).
208
ACTIVITIES: ELM FORK
1015
202530
3540
4550
TOOL HUNTING FOODPREPARATION
LITHIC SCATTER
SIT
E (C
OU
NTS
)
ARCHAIC LATE PREHISTORIC PREHISTORIC
ACTIVITIES: WEST FORK
10
11
12
13
14
15
16
TOOL HUNTING FOODPREPARATION
LITHIC SCATTER
SIT
E (
CO
UN
TS
)
ARCHAIC LATE PREHISTORIC PREHISTORIC
Figure 59. Prehistoric activities and time periods.
Relationships between the number of sites that contain evidence of each particular
activity is illustrated (figure 59). This data is biased by the number of surface sites that
contain limited evidence combined with the abundant data recovered from excavated
sites. It would appear that there is a decrease of activities from the Archaic to the Late
Prehistoric and very little lithic scatter during the Archaic along the Elm Fork. This is
209
not particularly the case. Lithic scatter, for example is a category that includes sites
that contained non-diagnostic lithic scatter. As a result, this category is composed of
non-diagnostic single component prehistoric sites. The same situation is true when
occupation frequency (figure 60) is considered. Future research with a better sample
would help solve this problem.
ACTIVITIES: ELM FORK
10
15
20
25
30
35
40
45
TOOL HUNTING FOODPREPARATION
LITHIC SCATTER
SIT
E (
CO
UN
TS)
MULTI-COMPONENT SINGLE-COMPONENT
ACTIVITIES: WEST FORK
10
11
12
13
14
15
16
17
TOOL HUNTING FOODPREPARATION
LITHIC SCATTER
SIT
E (
COU
NT
S)
MULTI-COMPONENT SINGLE-COMPONENT
Figure 60. Prehistoric activities and occupation frequency.
210
Most of the activities (figure 61) are consistently associated with water deposited
sediments and sandstone geological settings. Repeated testing indicates a preference
for these locations. Shale settings are followed by limestone. A similar pattern emerges
with respect to occupation density (figure 61). Many of the referenced site catchments
contain both sandstone and water deposited sediments such that overlap should be
taken into account.
GEOLOGICAL SETTING: ACTIVITIES
10
20
30
40
50
60
WATERDEPOSITEDSEDIMENTS
LIMESTONESETTINGS
SHALESETTINGS
CLAY RICHLIMESTONESETTINGS
SANDSTONESETTINGS
SIT
ES (
COU
NTS
)
TOOL HUNTING FOOD LITHICS
GEOLOGICAL SETTING SITE: DENSITY
10203040506070
WATERDEPOSITEDSEDIMENTS
LIMESTONESETTINGS
SHALESETTINGS
CLAY RICHLIMESTONESETTINGS
SANDSTONESETTINGS
SIT
ES
(CO
UN
TS)
EPHEMERAL MODERATE INTENSE
Figure 61. Prehistoric activities, site density and geological settings.
211
Patterns related to soil settings (figure 62) indicate a fairly equal distribution of soil
types within catchments that contain various activities and degrees of intensity. The
change between categories appears to be primarily related to the number of sites in
each category. The presence of the various soil settings indicates a strong correlation
with ecotonal settings.
SOIL SETTING: ACTIVITIES
10
20
30
40
50
60
CLAY RICHENVIRONMENTS
SANDYENVIRONMENTS
LOAMYENVIRONMENTS
OTHER
SIT
ES
(C
OU
NTS
)
TOOL HUNTING FOOD LITHICS
SOIL SETTING: DENSITY
102030405060708090
CLAY RICHENVIRONMENTS
SANDYENVIRONMENTS
LOAMYENVIRONMENTS
OTHER
SITE
S (C
OU
NTS
)
EPHEMERAL MODERATE INTENSE
Figure 62. Prehistoric activities, site density and soil settings.
212
great groups (figure 63) tend to produce similar patterns in relation to activities and site
density as did soil settings. The fairly equal distribution appears to be proportionately
affected by the number of sites within each category. In addition, the presence of all
three great groups supports ecotonal utilization regardless of site density or activity.
GREAT GROUPS: ACTIVITIES
10
20
30
40
50
60
MOLLISOLS VERTISOLS ALFISOLS ENTISOLS ANDINCEPTISOLS
SIT
E (C
OU
NT
S)
TOOL HUNTING FOOD LITHICS
GREAT GROUPS: SITE DENSITY
102030405060708090
MOLLISOLS VERTISOLS ALFISOLS ENTISOLS ANDINCEPTISOLS
SITE
(C
OU
NTS
)
EPHEMERAL MODERATE INTENSE
Figure 63. Prehistoric activities, site density and great groups.
213
Topographic settings (figure 64) in relation to activities and site density indicate a
variable landscape surrounding site location. Apparently sites are situated in locations
where the catchments include more then one topographic setting. The pattern
continues to proportionately change reflecting utilization of a variety of settings
regardless of activity or site density.
TOPOGRAPHIC SETTING: ACTIVITIES
10
20
30
40
50
60
BOTTOMLANDS WATER(DRAINAGES)
FLOODPLAINS UPLANDS TERRACES
SIT
ES (
COU
NTS
)
TOOL HUNTING FOOD LITHICS
TOPOGRAPHIC SETTING: SITE DENSITY
1020304050607080
BOTTOMLANDS WATER(DRAINAGES)
FLOODPLAINS UPLANDS TERRACES
SIT
ES (
COU
NTS
)
EPHEMERAL MODERATE INTENSE
Figure 64. Prehistoric activities, site density and topographic settings.
214
Given the previous results the ecotonal physiographic setting (figure 65) should have a
higher site frequency. However this graph is a little misleading due to the fact that
ecotonal settings contain a combination of the other categories. This cannot be
avoided due to the site catchment character combined with site distribution. Future
research should delineate sites that are completely within prairie or woodland settings.
The results would probably remain the same due to the character of the region.
PHYSIOGRAPHIC SETTING: ELM FORK
10
1520
2530
3540
45
PRAIRIES CROSSTIMBERS
MAJORRIPARIAN
CORRIDORS
MINORRIPARIAN
CORRIDORS
ECOTONE
SIT
E (
CO
UN
TS)
TOOL HUNTING FOOD LITHICS
PHYSIOGRAPHIC SETTING: WEST FORK
10
20
30
40
50
60
70
PRAIRIES CROSSTIMBERS
MAJORRIPARIAN
CORRIDORS
MINORRIPARIAN
CORRIDORS
ECOTONE
SIT
E (
CO
UN
TS)
EPHEMERAL MODERATE INTENSE
Figure 65. Prehistoric activities, site density and physiographic settings.
215
Statistical Conclusions Results yielded from the statistical analysis indicate that differences between site
locations through time are predominantly due to chance. Similarly there is no indication
of differential selection in relation to occupation frequency. Apparently prehistoric
populations in north-central Texas returned to similar site settings that may vary in
geographical location but not in economic related content. It is important to note that
the statistical results are extremely biased due to the number of sites situated Elm Fork
and the lack of sites along the West Fork of the Trinity. The bias is such that a robust
statistical comparison is not possible until more sites are uncovered. Furthermore,
repeated occupation at sites through time along the Elm Fork heavily weighted
temporal considerations and occupation frequency. The raw environmental and
economic data indicates differential selection. Maps depict this variability in relation to
site catchments. The particular variation between site locations is primarily due to the
different soil settings so there is a considerable difference between sites situated on the
Elm Fork, East Fork and West Fork of the Trinity.
The statistical results (table 8-table 10) are summarized to illustrate the variables
that were depicted and the corresponding results. Although statistical results indicate
that there is not a significant difference between site catchments in relation to
occupation frequency, density, activities or change through time there are a few cases
that do indicate statistically significant differences. Activities and site density indicate
statistically significant differences in relation to change through time and occupation
frequency. This simply means that site function and density varies through time but
216
Table 8. Summary of statistical conclusions.
Variable 1 Variable 2
X-Square df
p-value Statistical
Significance Great groups Components 22.1639 33 0.9239 N Great groups Time 26.2866 84 1 N Great groups C-Time 9.2734 48 1 N Great groups Site Density 13.0649 24 0.965 N Great groups Site Function 13.4503 36 0.9998 N Topography Components 40.1809 45 0.6759 Y Topography Time 49.3356 105 1 N Topography C-Time 19.1622 60 1 N Topography Site Density 19.7875 30 0.922 N Topography Site Function 15.5873 45 1 N Soil Setting Components 60.4014 60 0.4612 Y Soil Setting Time 62.8236 140 1 N Soil Setting C-Time 20.3485 80 1 N Soil Setting Site Density 26.7694 40 0.9459 N Soil Setting Site Function 17.6018 60 1 N Geology Components 34.2261 45 0.8789 N Geology Time 35.4769 105 1 N Geology C-Time 15.3905 60 1 N Geology Site Density 15.3535 30 0.9876 N Geology Site Function 12.4191 45 1 N Physiography Components 21.4053 36 0.9743 N Physiography Time 21.7192 84 1 N Physiography C-Time 8.007 48 1 N Physiography Site Density 9.2363 24 0.997 N Physiography Site Function 8.2276 36 1 N Site Density Components 58.4496 6 0 Y Site Density Time 64.6353 14 0 Y Site Density C-Time 2.4361 8 0.9646 N Site Function Components 32.2334 9 0.0002 Y Site Function Time 41.2895 21 0.0052 Y Site Function C-Time 1.9654 12 0.9995 N Elm Fork and West Fork Combined
not necessarily in relation to site location. Occupation frequency exhibits statistically
significant differences in relation to topography and soil setting. This is the only
indication of differences between site catchments and given the character of the
variables this suggests that there is a pattern operating that can be correlated with site
217
Table 9. Summary of statistical conclusions.
Variable 1 Variable 2 X-Square df p-value ss great groups Sing/Multi (West) 0.5632 3 0.9048 N great groups Sing/Multi (Elm) 1.2972 3 0.9775 N great groups Sing/Multi (Total) 3.3307 11 0.9856 N Topography Sing/Multi (West) 0.6025 4 0.9628 N Topography Sing/Multi (Elm) 0.9917 4 0.911 N Topography Sing/Multi (Total) 5.8985 14 0.969 N Soil Setting Sing/Multi (West) 3.9752 3 0.2642 Y Soil Setting Sing/Multi (Elm) 2.547 3 0.4669 Y Soil Setting Sing/Multi (Total) 5.7574 20 0.9992 N Geology Sing/Multi (West) 0.3855 4 0.9836 N Geology Sing/Multi (Elm) 1.0043 4 0.9091 N Geology Sing/Multi (Total) 2.985 15 0.9996 N Physiography Sing/Multi (West) 0.4115 4 0.9815 N Physiography Sing/Multi (Elm) 1.0354 4 0.9044 N Physiography Sing/Multi (Total) 2.4332 12 0.9984 N Site Density Sing/Multi (West) 4.9184 2 0.0855 Y Site Density Sing/Multi (Elm) 17.1127 2 0.0002 Y Site Density Sing/Multi (Total) 46.1361 2 0 Y Site Function Sing/Multi (West) 2.31 3 0.5106 Y Site Function Sing/Multi (Elm) 9.7762 3 0.0206 Y Site Function Sing/Multi (Total) 38.7188 3 0 Y West and Elm Fork Individually Analyzed
reoccupation. Of all of the variables these two are probably the most informative in
relation to extending interpretation beyond the physical landscape. Along the Elm Fork
of the Trinity there is a statistically significant difference between soil settings with
respect to change through time. Essentially, this means that sites were situated in
areas that would provide access to specific types of vegetation. In addition, there is a
statistically significant difference in site density in relation to soil setting.
218
Table 10. Summary of statistical conclusions.
This i
Setting
there
time in
that
Variable 1 Variable 2 X-Square df p-value ss great groups A/LP/P (West) 1.4139 6 0.965 N great groups A/LP/P (Elm) 1.7195 6 0.9436 N Topography A/LP/P (West) 2.209 8 0.9739 N Topography A/LP/P (Elm) 1.7437 8 0.9879 N Soil Setting A/LP/P (West) 0.257 6 0.9997 N Soil Setting A/LP/P (Elm) 4.0219 6 0.6737 Y Geology A/LP/P (West) 1.3967 8 0.9943 N Geology A/LP/P (Elm) 3.211 8 0.9204 N Physiography A/LP/P (West) 1.4773 8 0.9931 N Physiography A/LP/P (Elm) 1.0117 8 0.9982 N Activities A/LP/P (West) 1.9469 6 0.9245 N Activities A/LP/P (Elm) 49.7815 6 0 Y Activities Soil Setting 1.9554 9 0.9922 N Activities Topography 3.6075 12 0.9895 N Activities Geology 4.5979 12 0.9707 N Activities Physiography 4.55 12 0.9713 N Activities great groups 2.4048 9 0.9833 N Density Soil Setting 3.1109 6 0.7948 Y Density Topography 3.3872 8 0.9078 N Density Geology 4.4548 8 0.8139 N Density Physiography 3.658 8 0.8866 N Density A/LP/P (West) 5.1097 4 0.2762 Y Density A/LP/P (Elm) 67.3799 4 0 Y Density great groups 3.6117 6 0.7291 Y West and Elm Fork Individually Analyzed
ndicates that site density varies according to the particular soil setting. Soil
s tie directly into the edible, medicinal and supplemental vegetation given that
is an association between the two. The fact that site density and change through
correlation with soils settings produce statistically significant results is a signal
prehistoric populations were drawn to specific locations related to the
219
environmental setting. Future hypotheses could be formulated to determine what types
of plants these cultures were drawn to during their respective time periods.
Topographic location provides valuable information in relation to climatic
fluctuations and micro-environmental variation. The statistical significance in this
respect indicates that patterns exist between site location and topographic setting. As
these settings are evaluated it is possible to form explanations for site location in
relation to the climate. As such, this variation is addressed in the final synthesis.
Overall, it can be generally inferred that prehistoric populations consistently selected
site locations for intensive or ephemeral occupation that vary according to soil setting
and topographic setting. These are important clues to discovering the motivation for
such choices. Due to the sample bias posed by the numerous sites along the Elm Fork
that are in close proximity or indicate reoccupation it is necessary to approach the
development of the regional settlement pattern model from a different angle.
One of the benefits of incorporating GIS into SCA is the ability to spatially analyze
site locations with overlay analysis. The GIS interface provides a means to observe the
intersection of the multiple variables, interpret these patterns into dynamic relationships
and observe change through time. In the final synthesis, prehistoric settlement
patterns are observed comparatively between the East, West and Elm Forks of the
Trinity as they change through time by performing an overlay analysis with a GIS
interface. Results from this analysis combined with tabular and descriptive data are
synthesized in a regional model of prehistoric settlement pattern behavior in north-
central Texas.
220
CHAPTER 7
NORTH CENTRAL TEXAS SETTLEMENT PATTERN MODEL
The model presented here is designed to address the character of prehistoric
adaptation to the environment and landscape of north-central Texas. A variety of
factors influence prehistoric decision-making processes. Each of these factors has been
evaluated throughout the course of this thesis. In order to make sense of the pieces
they must be integrated into a coherent whole and synthesized into a regional
settlement model. The primary goal of this research is to formulate theories that would
explain why these populations selected certain locations during their respective time
periods and why these locations were revisited during later time periods. Examination
of the landscape has helped determine the economic value of areas directly associated
with archaeological sites. Evaluation of archaeological remains at these sites has
delineated functional value of these locations. In this chapter sites are comparatively
evaluated and settlement models are developed for each time period within the West
Fork, Elm Fork and East Fork. Settlement patterns that emerge from these
relationships are then evaluated on a regional scale to develop theories regarding
functional site types and adaptations that characterize prehistoric populations. This
model can be utilized to formulate future hypotheses concerning relationships between
these cultures, how they related to each other, to the animals and to the landscape.
Prehistoric Populations along the West Fork of the Trinity
During the Early Archaic sites, along the West Fork of the Trinity, contain evidence
that indicates moderate occupation involving hunting in upland alluvial settings within
221
the Western Cross Timbers. The most intensively occupied site in this portion of the
study area (41WS38) is centrally located in relation to the other sites in the West Fork
drainage. Intensive occupation at this site is indicated by high artifact density, hearth
features, fire cracked rock and faunal remains that have been excavated from midden
deposits. A site of this nature is considered to be a semi-sedentary base camp that was
probably seasonally occupied. The proximity of this multi-component site to the rolling
plains suggests that it was a good location for bands to aggregate and prepare for
communal hunting on the plains. Of all the currently known sites in Wise County this
site appears to be the most frequently occupied based on the diversity of time-
diagnostic projectile points that have been recovered. This site is situated in an
economically advantageous position on the floodplain near a minor tributary. It is
located in a diverse location within the parameters of the Grand Prairie/Western Cross
Timbers ecotone. Economic values associated with this ecotonal locale indicate an
equal presence of edible and medicinal vegetation.
All of the sites that were occupied during the Early Archaic in the West Fork
drainage are also associated with the Middle Archaic time period. An additional site
location is present approximately 30 km to the north in a similar setting. It is located in
an ecotonal setting between the Western Cross Timbers and the Grand Prairie on top of
the same geological foundation but it is on a terrace near a major tributary (Denton
Creek) rather than on a floodplain of a minor tributary. This site is not as intensively
occupied but it has an equivalent economic value. Limited food preparation and
hunting activities are indicated by the artifacts recovered from this site. However, there
222
is no evidence of tool manufacture. Economic potentials are equally distributed
between edible, medicinal and supplemental values. Combined with the frequency of
peripheral non-diagnostic prehistoric sites it is suggested that the Howard site was a
base camp and 41WS7 was an ephemeral hunting camp.
Late Archaic patterns reveal continued site reoccupation of the Middle Archaic sites
with one new site established approximately 30 km to the west and a couple more
about 10km south of 41WS38. Site 41WS43 is situated in a terrace setting on top of
Antlers sandstone near the ecotone that connects the Western Cross Timbers to the
Rolling Plains. The site is ephemeral containing evidence of hunting activities
suggested by projectile points. However, the economic resources offered by the
riparian and ecotonal settings are fairly substantial such that it is likely that it served as
a hunting camp. The overall pattern during this time period suggests a certain level of
sedentism combined with the development of exploratory hunting camps in relation to
the southern sites. Late Archaic sites in the West Fork Study area are less numerous
with relatively lower wildlife and economic values than the previous time periods. The
two base camps are situated near the Grand Prairie/Western Cross Timbers ecotone
whereas the ephemeral hunting camps are in the interior of the Cross Timbers
approximately 5km from the Rolling Plains. This configuration suggests a western
movement from the base camps toward the rolling plains where large game could be
hunted. A westward shift of site locations is the most obvious between the Middle and
Late Archaic throughout the region. It is less noticeable in the West Fork region due to
the paucity of Middle and Late Archaic sites but when viewed in relation to the West
223
Fork this shift is more apparent. Presence of diagnostic points and darts at these
western sites combined with repeated occupation suggests either a migratory route or a
satellite hunting camp. Supporting evidence for a migratory route is provided by an
ephemeral prehistoric site containing non-diagnostic lithic scatter approximately 10 km
west of the nearest diagnostic site and well within the rolling plains. In addition,
although archaeological sites discovered at Lake Bridgeport, 3 km west of this site, have
not been reported to TARL, a multitude of evidence, housed at the University of North
Texas, awaits diagnosis. Sites in Wise County that are in the process of analysis could
add a greater breadth and meaning to this settlement model. Based on the current
settlement model it is expected that sites are located directly west of the study area
and that archaeological evidence reflects cultural affinities to other sites in the region.
Late Prehistoric sites in Wise County are few in number and exhibit opposite
patterns of the previous Late Archaic time period. One could hypothesize based on the
settlement patterns that emerged during this study that population density decreased
along with territorial parameters. During the first phase of the late prehistoric there is
only one site in the West Fork study area that contains evidence associated directly with
this time period designation. Only three sites are associated with the second phase of
the Late Prehistoric and two with the general Late Prehistoric. All of these sites are
occurrences of reoccupation of sites related to previous time periods. As such, it is
difficult to draw any certain conclusions about late Prehistoric populations in the West
Fork study area. However, it would suffice to say that these populations appear to be
retreating from the westward movement indicated during the Late Archaic.
224
Twelve prehistoric sites in Wise County are not associated with any given time
period due to the lack of time-diagnostic artifacts at these locations. However, the
presence of lithic scatter, tools and a variety of other materials that indicate prehistoric
human activity accumulate more meaning when combined with the character of the
landscape. All of these sites with the exception of the southernmost location are
associated with significantly high potential for openland and rangeland wildlife. In
addition edibility medicinal and supplemental values are significantly high at most of
these locations. They are also widely distributed across the landscape unlike the
clustered patterns that emerge in both the Elm Fork and East Fork areas. Given that
the relative distance from 41WS38 ranges between 10km and 30km a return to base
camp within a one day time frame is not likely. If they are related it is likely they were
satellite camps related to specialized activity or migratory camps that facilitated
movement toward the rolling plains to hunt bison and other large game.
Prehistoric populations were widely dispersed in the West Fork portion of the study
area (figure 43). These camps appear to be strategically situated in order to optimize
procurement efforts. Future hypotheses based on this model and corresponding
methodology could predict locations based on the character of the landscape.
Discovery and analysis of additional sites is necessary to further enhance knowledge
concerning prehistoric settlement patterns in the region.
225
Figure 66. Prehistoric populations along the West Fork of the Trinity.
226
Prehistoric Populations along the Elm Fork of the Trinity
Archaeological research has been performed extensively along the Elm Fork of the
Trinity (figure 44). As such, this research is heavily weighted by the increased number
of sites in this portion of the study area. However, it is worth noting that this area
provided a desirable habitat for prehistoric adaptations given the bottomland resources,
mast crop, access to permanent major water resources, ecotonal settings and the
overall diversity of the landscape.
Along the Elm Fork of the Trinity 11 of the 19 Early Archaic sites are aggregated in
a fairly central location along Little Elm Creek. Situated along the northern portion of
the Elm Fork near the Little Elm Creek branch these sites are either on the Woodbine
sandstone formation or on fluvatile terrace deposits. All of the sites in this area are on
the interior of the ecotone between the Western Cross Timbers and the Blackland
Prairie. A high overall economic value indicates a wide range of opportunities including
immediate access to mast crops. Intensity of occupation suggests that Early Archaic
populations recognized this and aggregated in this location. Settlement patterns during
the Early Archaic reveal aggregation in “base camp” locations with hunting, gathering
and/or migratory camps located within 10km to the west and lithic manufacturing
camps 20 km to the south. A wide range of mobility is indicated during this time
period. Gathering activities are complimented in the immediate vicinity of the sites and
the topographic setting is particularly advantageous during the wetter climatic episodes
that occurred during the Early Archaic. All of these factors indicate strategic site
selection.
227
A small aggregate of Early Archaic sites are equally spaced with approximately 1 km
separating them from one another. Collectively these sites form a westward facing
arch that appears to be protecting the confluence of the Elm Fork and Little Elm Creek
from Eastern invaders. These sites also follow the curvature of Running Branch Creek
providing immediate access to riparian resources. The strategic geographical location
of these sites is echoed by their functional presence that alternates between ephemeral
“outposts” and large base camps. These outposts could either signify an off-site
hunting and/or gathering event or sites protecting from northern or eastern
encroachments. On the other hand it is possible that rather than protecting the
confluence to the west these populations were simply following the river system
northward with scouts inspecting the territory upstream. North of these sites near the
confluence of Little Elm Creek a group of sites is situated in a heavily forested
environment in riparian environments. Of all of the sites in the region this group has
the highest frequency of reoccupation and density of artifacts.
There is an interesting inverse relationship between economic value and projectile
point frequency. Two sites in particular in the periphery of the Elm Fork illustrate this
relationship. For example, 41DN62 contains the largest number of artifacts yet a very
low economic value. Conversely, 41DN48 has the lowest number of artifacts and the
highest overall economic value. The wildlife factor is weighting this value although it
also the highest supplemental value and a fairly high edibility value. Hunting activities
are suggested by a combination of factors. The paucity of remains and the fact that it
only contains evidence of single occupation combined with the fact that all of the
228
elements necessary for the hunt including the appropriate habitat for rangeland wildlife
are present support a hunting situation. In addition a more intensively occupied site
(41DN31) is 1.5 km to the southwest that contains lithic debris, midden soil, fire
cracked rock and various other artifacts that suggests a relatively large scale gathering
covering approximately 10 acres. These factors collectively support an Early Archaic
gathering associated with a hunting event and camp near the confluence of Clear Creek
and the Elm Fork of the Trinity. Sherds and celts provide evidence of one other
occupation at site 41DN31 during the Late Prehistoric. Given the size and location of
the site it is very likely that Late Prehistoric people scavenged the site. There are three
reasons why this is thought to be the case. First, the contemporaneous occupation of
the two sites during the Early Archaic combined with the character and proximity of the
sites necessary for fulfillment of the conditions that typically facilitated a communal
hunt. Second, the fact that besides 41DN31 the closest Archaic sites are 7 km away
suggests that this scenario is likely. Third, the paucity of Late Prehistoric artifacts
combined with the size of the site and density of site content suggests that 41DN31
was scavenged for artifacts. The closest Late Prehistoric site is 1.4 km away on the
west side of the Trinity. This site (41DN2) situated on a terrace contains Perdiz,
Bonham and Fresno points, as well as, 20 ceramic sherds. It would be interesting to
compare the sherds from the two sites to evaluate possible similarities. Given the
tendency for territorial tribal conflict, especially during the Late Prehistoric, it is likely
that discovering a previously occupied site stimulated an increased level of
defensiveness. Therefore the selection of the site on a terrace with a river barrier
229
between the two sites appears to be a logical choice of location. This situation is a very
good example of how sites acquire meaning when viewed with respect to occupation
frequency and chronology in relation to other sites in the region.
A clustering of Early Archaic sites situated along the southern portion of Elm Fork
are correlated with fluvatile terrace deposits on the Blackland Prairie side of the Eastern
Cross Timbers ecotonal boundary. 41DN251 and 41DN252 contain Early Archaic
projectile points. These sites, situated in the Blackland Prairie near the confluence of
Denton Creek and the Elm Fork of the Trinity, are less than a kilometer from each
other. Both sites contain evidence of tool manufacture including cores, preforms,
debitage and diagnostic projectile points. Less than a kilometer to the south there are
three Archaic sites that contain non-diagnostic dart points, cores, debitage, preforms,
retouched pieces and fire cracked rock. These five sites are within a 1 kilometer radius
of each other. The closest economic resource is supplemental bottomland hardwoods
that would have provided fuel for the fire to heat treat lithic materials, provide warmth
and cook with. Situated on fluvatile terrace deposits within 2 km of a major tributary
these sites had access to the riparian food resources and protection from any
contemporaneous cultures traversing the River corridor. This cluster of sites appears to
be associated with preparation for the hunt or territorial warfare. A small number of
individuals present at this site are indicated by the moderate density of artifacts. More
than likely it was a band that was in the process of lithic manufacture while collecting
raw materials that were apparently available in the terrace settings of this portion of the
study area. Notably, they are the only Early Archaic sites along the Elm Fork that are
230
located in the Blackland Prairie proper. Once an occupation has been established at a
site it becomes difficult to discern which occupation had the strongest presence.
However, in situations such as this, the frequency of sites in proximity to one another
suggests that a significant amount of activity involving tool manufacture was conducted
in this area during the Early Archaic. The paucity of Late Archaic and late prehistoric
artifacts suggests that repeated visitation during later time periods was likely associated
with recycling of tools and raw materials. While this is not the case in all situations it
appears to be in the northern and southern satellite sites where economic values
associated with the landscape are relatively low within the site catchment. However,
both site clusters are situated near the Elm Fork of the Trinity and small patches of
edible vegetation. Medicinal and supplemental vegetation are also accessible although
not immediately present at the site. One of the sites contains a single Early Archaic
component with light scatter whereas the other site is dual-component and contains
substantial evidence of tool manufacture. Given that the sites are less then 1 km from
each other and contain similar diagnostic projectiles it is likely that there was cultural
interaction between these site locations or that a scavenging event occurred at a later
time. The site that contains considerably more lithic debitage contains evidence of
occupation during both the Early and Late Archaic.
During the Middle Archaic a cluster of sites are focused along Little Elm Creek in site
locations that were pre-established as base camps during the Early Archaic sites.
Besides this cluster there is only one other site located about 5 km south which is also a
case of reoccupation. Situated near the ecotone between the Eastern Cross Timbers
231
and the Blackland Prairie on fluvatile terrace deposits this site primarily contains
evidence of hunting activities that are expressed by a high density of projectile points.
However, the abundance and variety of projectile points indicates some type of tool
manufacture and the presence of cooking pits indicates food preparation. Bison
remains have also been recovered from the site in the form of scapula hoes. Given that
this site was occupied during the Early, Middle and Late Archaic time period it is difficult
to discern which archaeological remains can be attributed to each particular time
period. The main source of differentiation between the time periods is found in the
diagnostic points that are associated with stratified sites in peripheral locales.
During the Late Archaic the Elm Fork site location is once again re-occupied yet to a
much greater degree then previously. Site numbers in areas of reoccupation increase
substantially during the Late Archaic. Pre-established sites to the south and southwest
are also reoccupied. However, during this time period a number of new sites appear in
south, southwest, west and northwest locations. There also appears to be a trend of
movement to the west. For the most part new sites appear in small aggregates.
Choice locations appear to be fluvatile terrace deposits situated in sandstone settings.
All of the new site developments develop in these locations in or near ecotonal settings.
Newly developed sites indicate a notable shift from the Blackland Prairie/Eastern Cross
Timbers ecotone westward to the Grand Prairie/Eastern Cross Timbers ecotone. The
only exception is the southernmost sites situated approximately 5km south of the
camps that were established during the Early Archaic with new peripheral site locations
indicating occupation during the Late Archaic. Activities at these sites appear to be
232
centered on hunting and tool manufacture. 41DN68 and 41DN36 are the exceptions to
the overall pattern that is suggested by Late Archaic sites. These sites are situated
within the Eastern Cross Timbers in upland settings and on fluvatile terrace deposits.
While the overall economic values associated with these sites are fairly low the edibility
values are relatively high. Moving from east to west they are approximately 7 km from
the nearest site indicating that these sites were probably migratory camps. Given the
substantial increase of sites along the Elm Fork of the Trinity there appears to be a
significant population growth during this time period paired with a westward expansion.
Settlement patterns along the Elm Fork of the Trinity with respect to the diagnostic
sites containing Late Prehistoric remains are difficult to discern based on site
reoccupation. During the Late Prehistoric I time period site distribution returns to the
eastern edge of the Eastern Cross Timbers where an ecotone forms with the Blackland
Prairie. Terrace and upland settings in these locations continue to be utilized. During
the Late Prehistoric II time period identical settings are occupied. More information is
available from the distribution of sites that do not contain diagnostic projectiles but
contain untyped arrows and/or ceramics. These sites reveal a continued relationship
with the Grand Prairie/Eastern Cross Timbers eco-tonal boundary and fluvatile terrace
deposits. The ephemeral character of the sites combined with the variety and
abundance of valuable economic resources suggests a correlation with gathering
activities. During the Late Prehistoric populations appear to be growing yet retreating
to previously occupied sites and aggregating into smaller territorial eco-niches.
233
Figure 67. Prehistoric populations along the Elm Fork of the Trinity.
234
Prehistoric Populations along the East Fork of the Trinity
Extensive archaeological research has been conducted in the East Fork portion of
the study area (figure 45) which contains sites that are particularly unique in character.
Given the current state of digital data available for Collin County it was not possible to
evaluate the economic values. Geological information, great groups and topographic
settings are evaluated with respect to site selection. However, the variation is limited
based on the general uniformity of the landscape. All of the sites along the East Fork
are located in the center of the Blackland Prairie. Clay-rich soil characteristic of this
area involve a lot of shrink-swell activity that makes it very difficult for soils to develop.
As a result, besides the predominant tall grass prairie vegetation the main vegetal
resources of food, medicine and supplemental products are contained in the riparian
environments. Site distribution follows this pattern fairly closely.
Very little is known about the Archaic time period. Reportedly only a few sites
were occupied during the Early Archaic. These sites were locations that were later
selected for the construction of Wylie Focus Pits. Whether of not the Wylie Focus Pit
originated during the Early Archaic is unknown but by the Middle Archaic the sites that
would contain these characteristic ceremonial “pits” all provide evidence of prehistoric
activity. Late Archaic populations do not establish new sites but reoccupy sites
established during the Early Archaic. Perhaps these populations had a wider dispersal
across the landscape but were discovered in these locations due to the extensive
excavations that were conducted due to the presence of the Wylie Focus Pits that were
presumably not established until the Late Prehistoric.
235
While some sites suggest occupation during the Archaic time period the bulk of
archaeological remains such as the unusual “Wylie Focus Pit” depressions associated
with hearth features, lithic manufacture, evidence of violent deaths, burials and possible
ceremonial activity does not emerge until the late Prehistoric. This activity probably
corresponds with the increase of population and territoriality that would have put
prehistoric populations in a defensive stance. Supportive evidence for an increase in
populations during the Late Prehistoric is substantiated by the addition of 17 diagnostic
sites along the East Fork of the Trinity and 19 non-diagnostic sites scattered along the
minor tributaries to the east. In relation to the current settlement model significant
patterns emerge from site distribution in the East Fork portion of the study area. These
patterns relate to the centrally based camps that are equally spaced from each other
(8km-10km) across the landscape and then surrounded by clusters of subsidiary sites
that may or may not contain diagnostic material. Whether these sites were hunting
camps, gathering camps, migratory camps or lithic manufacturing camps remains to be
seen. Further research would shed light on the character of these subsidiary sites that
would either support or refute the current settlement model.
The overall character of the base camp sites, artifacts and number of burials
suggests that these sites were inhabited by completely different cultures than those
who inhabited the Elm Fork or West Fork. In addition, the causes of death indicate
conflict and likely conflict of territorial origin. What is difficult to discern is the direction
that the conflict is coming from. Given the character of the sites it seems that Eastern
populations were moving westward where they met with extreme resistance.
236
Figure 68. Prehistoric populations along the East Fork of the Trinity.
237
Functional Site Types in North-Central Texas
Regardless of time period there are common characteristics and functional
differences between sites that prompt the development site types. Five types of sites
are formulated as a result of this research and serve as the core for the current model
of prehistoric settlement patterns. These site types are referred to as base camps,
hunting camps, gathering camps, lithic manufacture camps and migratory camps.
Base camps should contain evidence of extended or multi-component occupation
and be situated in strategic locations that provide increased environmental diversity
with access to optimal economic resources to maximize the gathering contribution of
women and children. These sites should contain higher artifact densities and lithic
manufacture as it is not economic to transport large amounts of lithic materials during
the hunt. Tool manufacturing debris, food preparation artifacts, and food storage
containers are associated with base camps. Abundance of projectiles indicates work
stations at base camps.
Hunting camps are sites that contain evidence of ephemeral single visitation are
situated in strategic locations where hunting would be optimized. In particular, these
sites are associated with areas that have a higher potential for wildlife resources. As
such, these sites provide access to economic resources that may be more specialized in
terms of value relative to sites that contain evidence of higher density occupation. Sites
with low artifact density may have been utilized as food-processing stations where the
hunted animal was prepared for transport and dull or excess flakes discarded on site.
At these site locations there may also be evidence of tool maintenance, a logical by-
238
product required for the process of preparation. Moderate artifact diversities indicate
hunting camps depending on the character and abundance of the remains.
Gathering camps are sites that contain evidence of ephemeral or repeated
occupation combined with a low artifact density. This site type acquires meaning in
relation to the relative economic value. It is expected that at these particular sites
there is a high economic value for edible, medicinal, supplemental resources or a
combination therein. In order to be categorized as a camp the site location must have
access to a water source.
Lithic manufacture camps are sites that contain artifacts that are primarily and
directly associated with lithic manufacture. In addition, these sites if there are to be
seen as specialized activity sites will be located near areas that would provide
immediate access to lithic raw materials. It is also expected that such sites will contain
heat-treated or fire cracked rocks and in a location that has access to wood bearing
resources. While edible vegetation would be of secondary importance water must be
accessible within a reasonable distance from the site.
Migratory camps are ephemeral sites that contain similar archaeological evidence to
food procurement camps. This site type takes on meaning in association with other
sites that are similar in character and indicate a directional movement toward a territory
that would provide an economic resource that is seasonally available. In addition, these
sites will demonstrate a geographical or archaeological relationship with specific base
camps. While these sites are not expected to be associated with high economic values
they are expected to be situated in a location that has access to a water resource.
239
Prehistoric Adaptations
Prehistoric adaptations in north-central Texas vary according to the relative
geographical and ecological setting. Similarities and differences between the relative
settlement patterns have emerged during the course of this research. Factors such as
geology, soils, physiographic setting, topographic setting, climate, economic resource
base and population density influence the range of necessary adaptations.
Early and Middle Archaic cultures in north-central Texas set the stage and
established the foundations for future adaptations in this region. This is mainly due to
the natural circumstances they had to adapt to, as the landscape responded to the
changing conditions, during the transition from late glacial to post glacial climates.
Essentially, humans had to adjust to the establishment of extensive forests that
accompanied this transition. As such, it is expected that sites would be situated near
bottomland resources. This appears to be the case, particularly during the Early Archaic,
based on a combination of the faunal remains and geographic distribution of sites.
Faunal remains and the geographic distribution of sites also indicate that these low
density populations enjoyed a fairly wide range of mobility. However, sites associated
with the Early and Middle Archaic tend to aggregate in characteristic locations, with a
few exceptions, which can be explained by specific cultural adaptations. Lynott (1977)
observes a shift from open-land big game hunting to bottomland riverine hunting and
gathering during the Early and Middle Archaic. According to faunal evidence,
topographic and physiographic settings this appears to be the case. McCormick et al.
(1975) correlate Archaic seasonal camps with ecotonal settings and secondary activity-
240
specific sites spread out across the landscape strategically. It has been observed
during the course of this research that sites are primarily situated in ecotonal settings
with few exceptions. These exceptions find logical explanations by correlating the site
contents with the character of the landscape. For example, the correlation of lithic
manufacturing stations in association with geological settings, that would provide raw
materials, is a common situation.
On a regional scale something happened between the Middle and Late Archaic that
prompted a large scale westward expansion. The reason why a migration is
hypothesized is due to the character of the newly established. Sites that were
considered to be a part of this westward expansion were single occupation ephemeral
sites situated between 10-20 km from the “base camp” sites. The direction toward the
Plains is a probable migration route given that populations are known to have migrated
to the Plains during certain time periods to hunt bison. It is also known that trade with
the Plains Indians and New Mexico Pueblos occurred. Although, it is possible that
populations were expanding their site distribution away from the more desirable riverine
basins due to population increase, there is no evidence that indicates that they were
moving their base camps. Given the dry climatic conditions of the previous time period
and the wetter climate experienced by Late Archaic cultures it is likely that the
population increased and that they were expanding their hunting territory. There is
clearly an increase in the number of sites between the time periods which does suggest
an increase in population. Lynott (1977) observes an increase in population and
decrease in territoriality. Site distribution and frequency combined with the climate
241
patterns supports this. The wetter conditions provided more abundant resources so it is
not likely that they migrated out of the region. It is possible that they were returning to
the region after the Middle Archaic drought. The paucity of sites in the region during
the Middle Archaic suggests that populations were responding to drought conditions by
migrating to other locales so it is highly likely that they would have returned to their
previously established base camps when conditions permitted. The character of the
newly established sites is ephemeral and suggests hunting activities by the presence of
diagnostic point types. The complexity introduced by these sites is their distance from
any peripheral base camps and the appearance of all of these sites to the west of the
majority of established base camps. This is a question that can only be answered by
future research and the discovery of additional base camps.
Typically the initial pattern established by Archaic cultures involves a centrally
located semi-sedentary base camp surrounded by ephemeral camps that facilitated
hunting, gathering, migration or lithic manufacture. Besides the functional site types,
that appear to be re-established through time, regional behavior is discernible by the
geographical distribution of sites. Archaic cultures have a much wider range of
occupation then late prehistoric cultures. This is a phenomenon that requires more
research with respect to environmental and cultural resources. There is more than one
reason. Perhaps it is more straightforward for the late prehistoric but for the Archaic it
is not just a question of territoriality. It is a question that has to do with why Archaic
cultures chose to be highly mobile when they could have subsisted just as other
cultures did during other time periods under the same environmental conditions.
242
Late prehistoric cultures in north-central Texas walked upon common ground and
the knowledge passed down from their ancestors. These populations were drawn back
to the same locations that were utilized by the Archaic cultures. Base camps were re-
established and hunting/gathering camps were either revisited or re-established in
similar environments where economic values facilitated such activity. Lithic
manufacturing stations probably were established in locations where raw materials were
available naturally or where previous occupants left workable lithic debris behind. Tools
were likely recycled, replicated or refashioned. Cumulative knowledge concerning the
proper use of medicinal, edible and supplemental vegetation probably opened doors to
greater variety. However, paired with the knowledge passed on from previous
generations, population growth and pressure created new problems. Climatic
conditions would have influenced the success of the harvest of a seasonal plant or the
migratory patterns of animals that provided sustenance for these populations. During
the second phase of the Late Prehistoric the effects of drought conditions reappear
along with the utilization of bottomland resources, decrease in aquatic resources and
increase in the variety of fauna utilized during this time period. Soil character and
degree of development certainly provided variable opportunities as late prehistoric
populations turned to horticulture. In some cases the physiographic setting provided
more opportunities and in others restrictions or barriers. Topographic settings appear
to respond to climatic conditions and more pronounced territorial issues.
Regardless of territorialism, late prehistoric cultures still found ways to situate
themselves in their “preferred” locations for extended amounts of time. If they were
243
out of their “comfort zone” it does not appear that they were for very long. Even
though the population certainly increased, late prehistoric populations tend to occupy
sites that were previously occupied by the Archaic, with a tremendous decrease in
territorial range. However, it should be noted that there are many unknowns with the
number of “non-diagnostic” prehistoric sites present in the region. This pattern is
probably more related to the character of the research conducted in the region than it
is a real reflection of the behavior of late prehistoric cultures.
Regardless of the time period or climatic conditions prehistoric preferences in
relation to site selection appear to be fairly consistent through time. As Bousmann and
Verrett (1973) suggested fluviatile terraces often function to accommodate base camps.
Also, as McCormick et al. (1975) pointed out, specialized activity locations are often in
locations that are amenable to the activity. Several investigators emphasize the
importance of ecotonal settings. Ecotones are predominantly selected to provide access
to food and shelter. The importance of ecotonal settings is supported by the
archaeological faunal evidence and the placement of the sites on the landscapes.
Riparian environments and sites situated with reasonable access to water is often
combined with the benefits provided by the ecotonal setting.
During the course of this research interesting patterns and fascinating questions
have emerged. Interestingly, it is evident that through time based camps were re-
established and ephemeral camps distributed either in the near proximity of these base
camps or in a wider scattered dispersal around the heart of the base camp sites. In
these cases reoccupation of ephemeral sites appears to be based more on random
244
chance then on some type of learned pattern. Reoccupation can be better explained
with the addition of an interpretable landscape expanding the range of possibilities that
in combination with the material remains could help explain site selection. This
ecological approach needs to be expanded and refined with additional environmental
data collection and evaluation. Future research designed to address regional questions
such as reoccupation and variable dispersal is the only way that these patterns will
come to light.
A pattern of rhythmic dispersal and return appears to be the central organizational
mechanism underneath prehistoric human behavioral choices in north-central Texas.
Only future research can determine why this pattern exists and if indeed history repeats
itself through a natural process, some sort of implicit instinct, that revolves around the
need to sustain human life. As a result, one must ask oneself if what happened in the
past becomes a part of our inherited inherent knowledge that is utilized to survive in
the present. This is a question that I think archaeological research was initially
designed to answer but unless all of the parts of the whole are organized, preserved,
shared and understood this question will remain unanswered. With the knowledge
provided by the current regional settlement model and the accompanying methodology
it is possible to conduct future research to test a number of hypotheses that question
site function, distribution and large scale regional events such as the Middle to Late
Archaic westward expansion. These more specific hypotheses can provide clues that
when viewed collectively will eventually lead to answers for the larger more abstract
questions that science continually seeks to answer.
245
CHAPTER 8
DISCUSSION AND FUTURE RECOMMENDATIONS
Throughout the course of this research ecological, geographical and archaeological
methods have been evaluated, cultural histories reviewed and a new innovative
research methodology formulated to facilitate future research and enhance
interpretations of prehistoric behavior in north-central Texas. In order to build a
regional settlement model SCA has been conducted to extract and interpret
environmental information to shed light on the character of archaeological sites in
combination with previously excavated remains. Geographic Information System (GIS)
software has served as a useful tool for illustrating, comparing, organizing, storing and
explaining observations on a regional scale.
Archaeologists concerned with the dynamics of prehistoric behavior must take into
consideration the dynamics of the ecological system. Given the evidence available for
north-central Texas the best one can hope for on a regional scale is to develop an
understanding of the dynamics of the micro-environments around site locations,
correlate this information to the material remains, relate it to the larger network of sites
and draw inferences by comparison. Framing the archaeological remains in an
ecological context, addressing a series of research questions based on ecological and
archaeological data and analyzing spatial patterning to formulate a regional model is
the basis of the proposed methodology. The methodology can be refined, reshaped
and extended to fit the particular character of the region where it is employed. What is
of primary importance is that basic conditions are met to fulfill multiple research efforts.
246
Essentially, what this means is that site catchment analysis (SCA) and all that this
phrase implies should be incorporated into archaeological methodology.
Essentially, learning is the basis of a particular manifestation of cultural behavior. In
order to understand the mechanism of cultural selection, the process by which new
behaviors are taught, learned and retained should be investigated. For example, it
must be taken into consideration that food procurement techniques were much more
dependent upon geographical location during prehistoric times than they are today and
as such learning would have been as intricately intertwined with spatial location. The
dynamic character of this cultural process makes it very difficult to discern in the
archaeological record but the environment provides clues that can help understand the
larger picture. In addition, a deeper understanding of the resources available to ancient
populations has shed light on necessary decisions associated with that specific type of
environment. For example, many medicinal plants are highly toxic and potentially lethal
if not prepared in an appropriate manner. Therefore, a medicinal environment requires
a certain degree of knowledge of what plants to choose and how to prepare these
plants. Another instance of learning relates to cultural affinities related to style which
could either be due to trading goods or exchanging knowledge. On the other hand, it is
often the case that flint-knappers discovers a particular tool type and based on their
knowledge of tool manufacture replicate the item without outside instruction. As such,
it is very difficult to ascertain the basis of cultural affinity and to what extent it is a
result of primary or secondary case of information diffusion. In this case examination
of the artifacts in isolation could result in the possibility of either case. However, it is
247
possible to establish theories when other variables are factored into the equation.
Other artifacts, faunal remains, palynological remains and possibly structural remains all
provide clues that shed light on cultural patterns. If a solid correlation is established
between two particular sites it is possible to look for processes and reciprocity.
In addition to the complexity of social systems the complexity of ecosystems poses
a problem in and of itself in terms of obtaining a database that is powerful enough to
reconstruct past systemic interactions between humans and their landscape. Vayda
and Rappaport (1968: 495-496) caution against the application of an ecological
approach by emphasizing its limitations (cited in Geier, 1974: 47). Regardless of the
limitations that accompany any scientific inquiry into the past it is beneficial to begin
constructing databases that include consideration of the components of the ecosystem
and their potentiality to exist in the past. Although the task may seem insurmountable
it is a worthwhile venture that provides a more comprehensive view of the potentiality
of past cultural systems. It is true, as Vayda and Rappaport (1968) point out, that most
statements of socio-environmental interaction are hypotheses only. However, this is
the case with any other inference that archaeologists make about the past. The
archaeological record presents “knowns” and “unknowns.” It is what we do with the
“knowns” that will determine the depth and breadth of our learning experience.
Future Recommendations
Destruction of archaeological sites by agriculture, urban expansion, highway
construction, reservoir development projects and a variety of other federal, state, and
locally funded construction projects is a continual phenomenon. Mitigation of this
248
problem has been attempted by the passing of legislation such as the National
Environmental Policy Act of 1969. The inclusion of cultural resources as a significant
part of environmental impact assessments has provided archaeologists with the
opportunity to become involved in research and planning of land development to a
certain degree. Legislation protecting archaeological resources has produced massive
data collection in the archaeological community. Coupled with the lack of
communication between state and local agencies this situation has led to increased
fragmentation in the archaeological record.
Archaeology as a discipline has only been academically active for a little over 50
years. As a relatively young discipline, experimentation with research methodology and
excavation techniques is still in progress. In general, archaeology has come a long way
since the antiquarian days of collecting and early pre-conceived anthropological notions
of aboriginal cultures. Inter-disciplinary studies have contributed a wealth of
information to help explain prehistoric behaviors. Cultural ecology, ethnography and
ethnobotany, to name a few, have provided a deeper understanding and appreciation
for the necessary intellect required for survival during prehistoric times.
We no longer think of the pre-ceramic plant-collectors as a ragged and scruffy bands of nomads; instead they appear as a practiced and ingenious team of lay botanists who know how to wring the most out of a superficially bleak environment (Flannery, 1968:67).
A plethora of archaeological methods have been innovated since the inception of
archaeology as an academic discipline. These methods have provided focused efforts
and increased efficiency in terms of time management while contributing pertinent and
valuable information to the archaeological record. However, the associated data and
249
information becomes lost in the shuffle along with the exclusion of data from the site
context that could provide important data for other investigations. As a result, many
scientific inquiries become redundant procedures and comparability is limited to the
research methodology under which the initial excavations were conducted.
Archaeologists have been so focused on the particulars that the larger picture is either
obscured or over-generalized. While attempting to establish a regional model much
time has been consumed by re-entry of data and information. In addition, there is a
significant amount of data exclusion due to the unavailability of resource information.
It is very difficult for growth to occur in isolation. Archaeology should be a cumulative
discipline that is organized and accessible. Regardless of the situation in the
archaeological community land development continues and site reports accumulate in
inaccessible locations. What is lacking in this system of piecemeal cultural resource
management efforts is a standardized methodology to guide local archaeological
societies, academic institutions, archaeological consultants and consulting firms. While
these entities are due creative process, there needs to be an established set of
minimum requirements in place for data collection and reporting, in order for
excavation efforts to benefit future research and facilitate comparability. In addition,
there needs to be a mechanism in place to centralize and organize these data for future
investigations. Growth in technology over the past five years has provided the means
to make this happen. GIS technology provides the perfect organizing mechanism that
is well suited to the character of archaeological data. The ultimate goal is the creation
and maintenance of centralized regional databases, delineated by physiographic
250
boundaries, that links the archaeological data to spatial location and interpretive
information. The construction of a database of this nature can facilitate a variety of
archaeological, statistical and spatial analyses. With this in mind, the current research
was designed to create a proto-type of a regional archaeological database that was
incorporated into a GIS interface and linked to an interpretive environmental database.
SCA was the established research methodology selected to create a model of prehistoric
settlement patterns in north-central Texas. The design is such that multiple analyses
could be performed and various research methodologies employed. At this point in
time there is a necessity to organize the data that has been collected so that it can be
accessed and synthesized for cumulative analyses. In order to fulfill these goals a
mechanism must be in place to organize our prior knowledge so that it can be used
analytically. Otherwise we will continue to piece together a past that is biased by the
available data that has been collected to fulfill particular research methodologies and
the record will remain as fragmented in reports as it is in the ground. There is
assuredly data that is irretrievable. However, simply organizing what we already have
and establishing standards for future data collection is a step in the right direction.
Future excavations should include analyses beyond the archaeological remains including
active field surveys and excavation of environmental information.
1. Systematic excavation of the site with particular attention to the recovery of economic data
2. Identification of floral and faunal species, and soil character, recovered from
cultural deposits
3. Delineation and description of the microenvironmental habitat of each species
251
4. Field survey of the environment in the region of the site including topographic features, available resources, and cultural limitations of access within a 1-2 km radius of the site location. This would require extensive field survey, test units to collect natural pollen and micro-faunal samples with stratigraphic time- depth, soil samples and vegetation samples of the modern natural environment.
5. Delineation of concentric temporal contours around the site representing travel
times from the site to facilitate interpretation and field strategy 6. Description of technological aspects of the culture represented at the site, with
particular reference to patterns of subsistence and settlement based on the assemblage character.
7. Identification of the number and proportion of the resources found in the deposit
which could have been obtained from each of the several zones as defined by the temporal contours
8. Determining the zone (site territory) which can account for procurement of the
respective resources.
During the Stone Age, Bronze Age, Iron Age, Medieval, Renaissance and the Industrial
age in the Old World, archaeological evidence suggests that hunter-gatherers in north-
central Texas enjoyed the comforts of a landscape structured only by plants, animals,
and other humans. For over twelve thousand years people lived in this region and the
only traces of their existence are in the archaeological record, as artifacts, features, and
concentrations of other types of archaeological materials. Without these records, from
our perspective, in time and space, they did not exist. The challenge, then, is to record
the existence of these prehistoric populations with as much detail as possible so that we
can learn from the behavior of our ancestors to create a better future. We live in an
Information age where data surpasses our current ability to deal with it adequately but
we can begin fashioning and structuring databases for the archaeological record with
integrated systems that will archive this information for future generations.
252
APPENDIX
253
Figure A1. Wildlife potentials and prehistoric sites.
254
Figure A2. Wildlife potentials and Archaic projectile points.
255
Figure A3. Wildlife potentials and Early Archaic projectile points.
256
Figure A4. Wildlife potentials and Middle Archaic projectile points.
257
Figure A5. Wildlife potentials and Late Archaic projectile points.
258
Figure A6. Wildlife potentials and Late Prehistoric projectile points.
259
Figure A7. Wildlife potentials and Late Prehistoric I projectile points.
260
Figure A8. Wildlife potentials and Late Prehistoric II projectile points.
261
Figure A9. Vegetation potentials and prehistoric sites.
262
Figure A10. Vegetation potentials and Archaic projectile points.
263
Figure A11. Vegetation potentials and Early Archaic projectile points.
264
Figure A12. Vegetation potentials and Middle Archaic projectile points.
265
Figure A13. Vegetation potentials and Late Archaic projectile points.
266
Figure A14. Vegetation potentials and Late Prehistoric projectile points.
267
Figure A15. Vegetation potentials and Late Prehistoric I projectile points.
268
Figure A16. Vegetation potentials and Late Prehistoric II projectile points.
269
Figure A17. Economic potentials and prehistoric sites.
270
Figure A18. Economic potentials and Archaic projectile points.
271
Figure A19. Economic potential and Early Archaic projectile points.
272
Figure A20. Economic potentials and Middle Archaic projectile points.
273
Figure A21. Economic potentials and Late Archaic projectile points.
274
Figure A22. Economic potentials and Late Prehistoric projectile points.
275
Figure A23. Economic potentials and Late Prehistoric I projectile points.
276
Figure A24. Economic potentials and Late Prehistoric II projectile points.
277
Table A1. Raw data for hypothesis testing.
EAST FORK OCCUPATION FREQUENCY CHANGE THROUGH TIMEGREAT GROUPS SINGLE DUAL TRI MULTI P A EA MA LA LP LPI LPII CHROMUDERTS 44 2 1 1 19 13 1 2 1 18 3 16 CHROMUSTERTS 27 0 1 0 13 6 0 1 0 10 1 9 HAPLAQUOLLS 37 2 1 1 16 10 1 2 1 17 3 15 HAPLUSTOLLS 32 2 1 2 12 13 1 3 1 15 4 13 OCHRAQUALFS 17 1 0 0 11 3 0 0 0 3 0 3 PALEUSTALFS 1 0 0 0 1 0 1 0 0 0 0 0 PELLUSTERTS 45 2 1 3 17 16 2 4 2 2 5 20 RENDOLLS 1 0 0 0 0 0 0 0 0 1 0 1 USTOCHREPTS 35 2 1 3 14 14 2 4 2 17 5 15 TOTAL 47 2 1 3 19 16 2 3 2 22 5 20 OCCUPATION FREQUENCY CHANGE THROUGH TIMETOPOGRAPHIC SETTING SINGLE DUAL TRI MULTI P A EA MA LA LP LPI LPII ALLUVIAL TERRACES 45 2 1 3 17 16 2 4 2 22 5 20 FLOODPLAINS 37 2 1 1 16 10 1 2 1 17 3 15 UPLANDS 44 2 1 1 19 13 1 2 1 18 3 16 UPLANDS, STREAM TERRACES 32 2 1 3 10 14 2 4 2 18 5 16 TOTAL 47 2 1 3 19 16 2 3 2 22 5 20 OCCUPATION FREQUENCY CHANGE THROUGH TIMEGEOLOGICAL SETTING SINGLE DUAL TRI MULTI P A EA MA LA LP LPI LPII ALLUVIUM 36 2 1 1 15 9 1 2 1 18 3 16 AUSTIN GROUP 5 0 0 0 0 2 0 0 0 4 0 4 FLUVATILE TERRACE DEPOSITS 21 1 0 3 9 9 2 3 2 9 3 9 MARLBROOK MARL 4 0 0 0 4 0 0 0 0 0 0 0 OZAN FORMATION 26 2 1 0 9 7 0 1 0 14 2 12 PECAN GAP CHALK 10 0 0 0 10 0 0 0 0 0 0 0 WOLFE CITY SAND 7 0 0 0 7 0 0 0 0 0 0 0 WATER 14 0 0 2 7 5 1 2 1 6 2 6 TOTAL 47 2 1 3 19 16 2 3 2 22 5 20 OCCUPATION FREQUENCY CHANGE THROUGH TIMEPHYSIOGRAPHIC REGION SINGLE DUAL TRI MULTI P A EA MA LA LP LPI LPII BLACKLAND PRAIRIE 47 2 1 3 19 16 2 3 2 22 5 20 RIPARIAN ZONE (EAST FORK) (2KM) 13 0 0 1 0 6 1 1 1 9 1 9 RIPARIAN ZONE (MINOR TRIB 2KM) 24 2 1 1 8 7 1 2 1 15 3 13 EPHEMERAL STREAMS (1KM) 13 0 1 0 9 2 0 1 0 4 1 3 TOTAL 47 2 1 3 19 16 2 3 2 22 5 20
278
Table A2. Raw data for hypothesis testing.
ELM/WEST FORKS OCCUPATION FREQUENCY CHANGE THROUGH TIMEGREAT GROUP SING DUAL TRI MULTI P A EA MA LA LP LPI LPII ARGIUSTOLLS 5 2 1 1 4 2 1 2 4 2 2 2 CALCIUSTOLLS 28 3 5 5 21 9 6 6 15 11 7 6 HAPLUDERTS 31 10 4 3 29 10 7 3 13 15 4 4 HAPLUSTALFS 57 13 7 12 46 21 17 10 27 29 12 17 HAPLUSTEPTS 39 17 4 10 33 16 14 9 23 28 11 13 HAPLUSTERTS 48 11 8 10 38 18 16 8 24 28 12 15 HAPLUSTOLLS 45 14 6 11 42 17 13 9 26 27 12 15 PALEUSTALFS 70 20 8 12 58 26 19 10 33 35 12 17 PALEUSTOLLS 15 2 1 0 12 3 2 0 2 2 0 1 RHODUSTALFS 3 0 0 0 3 0 0 0 0 0 0 0 USTIFLUVENTS 34 11 4 5 28 16 10 4 16 16 2 8 TOPOGRAPHIC SETTING SING DUAL TRI MULTI P A EA MA LA LP LPI LPII BOTTOMLANDS 26 9 4 4 20 14 9 3 14 15 2 6 FLOODPLAINS 55 17 7 12 47 22 16 10 30 31 12 17 LOW HILLS AND RIDGES 50 11 8 9 36 19 14 6 26 30 11 13 OLD TERRACES AND VALLEY FILLS 2 1 0 1 2 1 1 2 2 1 1 1 PLAINS 6 3 0 2 6 3 2 3 4 2 1 3 PLEISTOCENE AGE TERRACES 27 4 7 8 17 13 11 6 18 21 10 12 RIDGES 11 2 0 3 11 3 3 4 4 4 1 3 STREAM DIVIDES 9 0 0 1 9 1 0 0 1 1 1 1 STREAM TERRACES 57 14 7 11 48 22 15 10 26 27 11 16 TERRACES 41 7 6 8 30 13 12 6 20 22 10 12 UPLAND RIDGES 13 1 0 3 13 3 2 3 4 4 2 2 UPLANDS 68 20 8 12 59 44 32 15 55 57 21 26 TERRACES, FOOTSLOPES, FANS 28 15 4 8 23 14 12 6 20 25 10 10 GEOLOGICAL SETTING SING DUAL TRI MULTI P A EA MA LA LP LPI LPII ALLUVIUM 40 11 7 8 33 19 13 7 23 22 7 11 ANTLERS SAND 9 2 0 0 9 1 0 1 2 0 0 0 BOKCHITO FORMATION 2 0 0 0 2 0 0 0 0 0 0 0 EAGLE FORD FORMATION 10 5 1 2 8 3 4 2 4 7 3 4 FLUVATILE TERRACE DEPOSITS 36 14 5 9 31 17 12 7 22 25 9 13 GLEN ROSE LIMESTONE 1 0 0 0 1 0 0 0 0 0 0 LIMESTONE AND CLAY 6 1 0 1 6 1 1 2 2 1 1 1 MARL AND LIMESTONE 2 0 0 1 2 1 0 0 1 1 1 1 KIAMICHI FORMATION 1 0 0 0 1 0 0 0 0 0 0 0 PALUXY SAND 1 0 0 2 1 2 2 2 2 3 1 1 PAWPAW FORMATION 4 0 0 0 4 0 0 0 0 0 0 0 SURFICIAL DEPOSITS 1 0 0 1 1 1 1 1 1 2 0 0
279
Table A3. Raw data for hypothesis testing.
ELM/WEST FORKS OCCUPATION FREQUENCY
CHANGE THROUGH TIMESOIL SETTING SING DUAL TRI M P A EA MA LA LP LPI LPII BLACKLAND 37 8 6 9 29 13 13 7 20 21 11 13 CLAY LOAM 43 17 5 9 35 17 14 8 23 28 11 13 CLAYEY BOTTOMLAND 36 11 7 6 30 15 11 6 21 21 7 8 CLAYPAN PRAIRIE 52 15 6 11 43 20 15 9 26 27 12 16 DEEP REDLAND 6 0 0 0 5 1 0 0 0 0 0 0 DEEP SAND 14 2 4 6 11 8 8 8 13 10 5 7 ERODED BLACKLAND 15 5 2 5 11 4 7 3 9 15 7 7 LOAMY BOTTOMLAND 56 17 8 12 48 24 17 10 31 32 12 17 LOAMY PRAIRIE 0 1 0 0 0 1 0 0 0 0 0 1 LOAMY SAND 24 6 4 8 17 12 10 7 15 18 6 12 LOW STONY HILL 2 0 0 0 2 0 0 0 0 0 0 0 REDLAND 4 1 1 0 3 0 0 1 3 1 1 0 ROCKY HILL 1 0 0 0 1 0 0 0 0 0 0 0 SANDSTONE 3 1 0 0 3 1 0 0 1 0 0 0 SANDY LOAM 69 20 8 12 57 26 19 10 33 35 12 17 SANDY 19 2 3 4 11 12 5 2 11 8 3 6 SHALLOW CLAY 2 1 0 2 2 2 1 1 2 2 1 3 SHALLOW 1 0 0 0 1 0 0 0 0 0 0 0 STEEP ADOBE 8 1 0 1 7 1 1 2 2 2 1 1 TIGHT SANDY LOAM 11 3 0 3 11 4 3 4 5 4 1 3 WATER 61 16 8 11 51 23 18 9 30 32 11 16
ELM/WEST FORKS OCCUPATION FREQUENCY
CHANGE THROUGH TIMEPHYSIOGRAPHIC REGION SING DUAL TRI M P A EA MA LA LP LPI LPII BLACKLAND PRAIRIE 13 7 1 0 14 4 3 1 4 2 1 1 EAST CROSS TIMBERS 49 11 8 9 36 19 14 6 26 29 11 13 GRAND PRAIRIE 22 4 3 2 17 7 3 2 9 10 2 3 ROLLING PLAINS 1 0 0 0 1 0 0 0 0 0 0 0 WEST CROSS TIMBERS 11 3 0 3 11 4 3 4 5 4 1 3 RIPARIAN ZONE (ELM FORK) (2K) 16 1 3 0 11 4 3 0 4 5 0 1 RIPARIAN ZONE (WEST FORK) (2K) 2 0 0 1 2 1 1 1 1 2 0 0 RIPARIAN ZONE (DENTON CREEK) (2K) 3 1 0 0 4 0 0 0 1 0 0 0 RIPARIAN ZONE (MINOR TRIBUTARIES) (2K) 28 4 4 7 19 12 9 5 16 19 8 8 EPHEMERAL STREAMS (1KM) 20 4 4 4 15 12 6 3 11 11 3 4 ECOTONE (B PRAIRIE/E CROSS TIMBERS) (2K) 5 0 1 0 3 1 1 1 2 1 1 0 ECOTONE (E CROSS TIMBERS/G PRAIRIE) (2K) 14 3 3 1 10 6 2 0 7 8 1 2 ECOTONE (G PRAIRIE/W CROSS TIMBERS) (2K) 6 1 0 1 6 1 1 2 2 1 1 1 TOTAL 71 16 8 11 59 26 19 10 33 35 12 17
280
Table A4. Raw data for hypothesis testing.
ELM/WEST FORKS OCCUPATION DENSITY
GREAT GROUP EPHEMERAL MODERATE INTENSIVE TOTAL ARGIUSTOLLS 5 2 2 9
CALCIUSTOLLS 25 6 10 41
HAPLUDERTS 29 12 7 48
HAPLUSTALFS 55 21 13 89
HAPLUSTEPTS 39 18 13 70
HAPLUSTERTS 45 18 14 77
HAPLUSTOLLS 43 19 14 76
PALEUSTALFS 67 28 15 110
PALEUSTOLLS 14 3 1 18
RHODUSTALFS 3 0 0 3
USTIFLUVENTS 30 18 6 54
TOPOGRAPHIC SETTING EPHEMERAL MODERATE INTENSIVE TOTAL BOTTOMLANDS 22 15 6 43
FLOODPLAINS 52 24 15 91
LOW HILLS AND RIDGES 48 18 12 78
OLD TERRACES AND VALLEY FILLS 2 1 1 4
PLAINS 6 4 1 11
PLEISTOCENE AGE TERRACES 26 9 11 46
RIDGES 10 4 2 16
STREAM DIVIDES 9 0 1 10
STREAM TERRACES 53 23 12 88
TERRACES 39 13 10 62
UPLAND RIDGES 12 2 3 17
UPLANDS 66 28 15 109
TERRACES, FOOTSLOPES, ALLUVIAL FANS 29 15 11 55
GEOLOGICAL SETTING EPHEMERAL MODERATE INTENSIVE TOTAL ALLUVIUM 35 19 12 66
ANTLERS SAND 9 2 0 11
BOKCHITO FORMATION (UNDIVIDED) 2 0 0 2
EAGLE FORD FORMATION 10 6 2 18
FLUVATILE TERRACE DEPOSITS 34 18 12 64
GOODLAND LIMESTONE AND WALNUT CLAY (UNDIVIDED) 6 1 1 8
GRAYSON MARL AND MAIN STREET LIMESTONE 2 0 1 3
PALUXY SAND 0 1 2 3
PAWPAW FORMATION 4 0 0 4
SURFICIAL DEPOSITS (UNDIVIDED) 0 1 1 2
TWIN MOUNTAINS FORMATION 1 2 0 3
WOODBINE FORMATION 39 18 11 68
281
Table A5. Raw data for hypothesis testing.
ELM/WEST FORKS OCCUPATION DENSITYSOIL SETTING EPHEMERAL MODERATE INTENSIVE TOTAL BLACKLAND 36 13 11 60
CLAY LOAM 42 19 13 74
CLAYEY BOTTOMLAND 32 16 12 60
CLAYPAN PRAIRIE 50 21 13 84
DEEP REDLAND 5 1 0 6
DEEP SAND 11 8 7 26
ERODED BLACKLAND 15 7 5 27
LOAMY BOTTOMLAND 53 25 15 93
LOAMY PRAIRIE 0 1 0 1
LOAMY SAND 23 12 7 42
LOW STONY HILL 2 0 0 2
REDLAND 4 1 1 6
ROCKY HILL 1 0 0 1
SANDSTONE 3 1 0 4
SANDY LOAM 66 28 15 109
SANDY 17 8 3 28
SHALLOW CLAY 2 2 1 5
SHALLOW 1 0 0 1
STEEP ADOBE 7 1 2 10
TIGHT SANDY LOAM 10 5 2 17
WATER 56 26 14 96
TOTAL SITES 68 28 15 111
ELM/WEST FORKS OCCUPATION DENSITYPHYSIOGRAPHIC REGION EPHEMERAL MODERATE INTENSIVE TOTAL
BLACKLAND PRAIRIE 10 5 1 16
EAST CROSS TIMBERS 39 14 10 63
GRAND PRAIRIE 8 4 2 14
ROLLING PLAINS 1 0 0 1
WEST CROSS TIMBERS 10 5 2 17
RIPARIAN ZONE (ELM FORK) (2KM) 14 4 2 20
RIPARIAN ZONE (WEST FORK) (2KM) 1 1 1 3
RIPARIAN ZONE (DENTON CREEK) (2KM) 3 1 0 4
RIPARIAN ZONE (MINOR TRIBUTARIES) (2KM) 28 8 8 44
EPHEMERAL STREAMS (1KM) 17 8 7 32
ECOTONE (BLACKLAND PRAIRIE/EAST CROSS TIMBERS) (2KM) 4 1 1 6
ECOTONE (EAST CROSS TIMBERS/GRAND PRAIRIE) (2KM) 12 6 3 21
ECOTONE (GRAND PRAIRIE/WEST CROSS TIMBERS) (2KM) 6 1 1 8
TOTAL 68 28 15 111
282
Table A6. Raw data for hypothesis testing.
ELM/WEST FORKS ACTIVITIESGREAT GROUP TOOL MANUFACTURE FOOD PREP HUNTING LITHIC SCATTER ARGIUSTOLLS 3 4 4 4 CALCIUSTOLLS 9 16 15 14 HAPLUDERTS 15 24 19 17 HAPLUSTALFS 18 34 35 38 HAPLUSTEPTS 18 35 32 23 HAPLUSTERTS 17 35 33 29 HAPLUSTOLLS 21 40 34 21 PALEUSTALFS 26 45 44 40 PALEUSTOLLS 7 7 3 11 RHODUSTALFS 1 1 0 1 USTIFLUVENTS 15 20 24 16 TOPOGRAPHIC SETTING TOOL MANUFACTURE FOOD PREP HUNTING LITHIC SCATTER BOTTOMLANDS 13 17 21 12 FLOODPLAINS 23 43 40 27 LOW HILLS AND RIDGES 16 29 31 30 OLD TERRACES/VALLEY FILLS 1 2 2 1 PLAINS 3 3 5 3 PLEISTOCENE AGE TERRACES 12 27 28 19 RIDGES 4 6 6 5 STREAM DIVIDES 3 0 1 8 STREAM TERRACES 20 36 36 33 TERRACES 14 24 25 31 UPLAND RIDGES 2 6 5 8 UPLANDS 26 45 44 39 TERRACES, FOOTSLOPES, FANS 14 28 27 19 GEOLOGICAL SETTING TOOL MANUFACTURE FOOD PREP HUNTING LITHIC SCATTER ALLUVIUM 21 31 31 18 ANTLERS SAND 0 4 2 5 BOKCHITO FORMATION 0 0 0 2 EAGLE FORD FORMATION 2 7 8 6 FLUVATILE TERRACE DEPOSITS 21 32 32 19 GOODLAND LIMESTONE AND WALNUT CLAY 1 4 2 3 GRAYSON MARL AND MAIN STREET LIMESTONE 0 1 1 2 PALUXY SAND 2 2 3 0 PAWPAW FORMATION 0 0 0 4 SURFICIAL DEPOSITS 1 1 2 0 TWIN MOUNTAINS FORMATION 1 1 2 0 WOODBINE FORMATION 14 26 30 23
283
Table A7. Raw data for hypothesis testing.
ELM/WEST FORKS ACTIVITIES
SOIL SETTINGTOOL MANUFACTURE
FOOD PREPARATION
HUNTING ACTIVITIES
LITHIC SCATTER
BLACKLAND 14 25 26 27 CLAY LOAM 19 36 32 24 CLAYEY BOTTOMLAND 21 29 29 17 CLAYPAN PRAIRIE 21 36 35 36 DEEP REDLAND 1 0 0 3 DEEP SAND 7 12 16 7 ERODED BLACKLAND 3 12 13 10 LOAMY BOTTOMLAND 23 43 40 27 LOAMY PRAIRIE 1 1 1 0 LOAMY SAND 7 19 21 11 LOW STONY HILL 1 1 0 0 REDLAND 2 3 2 2 ROCKY HILL 1 0 0 0 SANDSTONE 0 1 1 2 SANDY LOAM 26 45 44 39 SANDY 4 9 11 11 SHALLOW CLAY 1 2 3 1 STEEP ADOBE 2 6 3 2 TIGHT SANDY LOAM 4 6 7 5 ELM/WEST FORKS ACTIVITIES
PHYSIOGRAPHIC REGIONTOOL MANUFACTURE
FOOD PREPARATION
HUNTING ACTIVITIES
LITHIC SCATTER
BLACKLAND PRAIRIE 7 11 6 5 EAST CROSS TIMBERS 11 22 25 27 GRAND PRAIRIE 4 6 6 3 WEST CROSS TIMBERS 3 7 7 5 RIPARIAN ZONE (ELM FORK) (2KM) 5 8 5 4 RIPARIAN ZONE (WEST FORK) (2KM) 2 1 2 0 RIPARIAN ZONE (DENTON CREEK) (2KM) 0 1 1 2 RIPARIAN ZONE (MINOR TRIBUTARIES) (2KM) 12 17 18 19 EPHEMERAL STREAMS (1KM) 8 14 15 10 ECOTONE (BLACKLAND PRAIRIE/EAST CROSS TIMBERS) (2KM) 3 2 2 4 ECOTONE (EAST CROSS TIMBERS/GRAND PRAIRIE) (2KM) 6 9 8 6 ECOTONE (GRAND PRAIRIE/WEST CROSS TIMBERS) (2KM) 1 5 2 2
284
Table A8. Raw data for hypothesis testing.
ELM/WEST FORKS OCCUPATION FREQUENCY CHANGE THROUGH TIMEOCCUPATION DENSITY SINGLE DUAL TRI MULTI P A EA MA LA LP LPI LPII EPHEMERAL 64 4 0 0 51 4 2 0 5 10 0 0 MODERATE 6 15 2 4 6 15 8 3 16 11 2 9 INTENSIVE 1 1 5 8 3 7 9 5 13 14 10 8 ACTIVITIES TOOL MANUFACTURE 11 5 5 5 12 7 9 3 12 12 7 7 CORES HAMMERSTONES PREFORMS BIFACES DEBITAGE FOOD PREPARATION 17 14 5 10 18 15 14 6 20 24 12 14 TOOLS GROUNDSTONE CERAMICS ANIMAL BONES FCR HUNTING ACTIVITIES 8 16 8 12 9 21 17 8 30 26 12 17 DIAGNOSTIC POINTS LITHIC SCATTER 39 1 0 0 31 2 0 0 4 4 0 0 ISOLATED LITHIC DEBRIS
285
Table A9. Economic values: Vegetation scores for soil settings. Common Name Family Genus Species Edibility Medicinal Toxicity Forage Other Boxelder ACERACEAE Acer various 0% 0% 0% 0% 100% Yarrow ASTERACEAE Achillea millefolium L. 50% 50% 0% 0% 0% Canada garlic LILIACEAE Allium canadense 0% 0% 0% 0% 100% Drummond onion LILIACEAE Allium mutabile 0% 0% 0% 0% 100% Wild onion LILIACEAE Allium sp. various 0% 0% 0% 0% 100% Amaranth AMARANTHACEAE Amaranthus spinosis L. 100% 0% 0% 0% 0% Red-Root Rigweed AMARANTHACEAE Amaranthus retroflexus L. 100% 0% 0% 0% 0% Western ragweed ASTERACEAE Ambrosia various 50% 50% 0% 0% 0% Bluestem, big POACEAE Andropogon gerardii 0% 0% 0% 50% 50% Bluestem, sand POACEAE Andropogon various 0% 0% 0% 50% 50% Potato bean FABACEAE Apios americana 100% 0% 0% 0% 0% Threeawn, perennial POACEAE Aristida various 50% 0% 0% 50% 0% Threeawn, purple POACEAE Aristida various 0% 0% 0% 100% 0% Threeawn, Wright POACEAE Aristida various 0% 0% 0% 100% 0% Sagewort ASTERACEAE Artemisia various 50% 50% 0% 0% 0% Butterfly weed ASCLEPIADACEAE Asclepias tuberosa 10% 90% 0% 0% 0% Milkweed ASCLEPIADACEAE Asclepias verticillata L. 10% 90% 0% 0% 0% Whorled milkweed ASCLEPIADACEAE Asclepias verticillata 10% 90% 0% 0% 0% Green antelope horn ASCLEPIADACEAE Asclepias viridis 10% 90% 0% 0% 0% Heath aster ASTERACEAE Aster various 0% 0% 0% 0% 0% Bluestem, cane POACEAE Bothriochloa various 0% 0% 0% 100% 0% Bluestem, silver POACEAE Bothriochloa various 0% 0% 0% 100% 0% Grama, blue POACEAE Bouteloua various 0% 0% 0% 100% 0% Grama, hairy POACEAE Bouteloua various 0% 0% 0% 100% 0% Grama, sideoats POACEAE Bouteloua various 0% 0% 0% 100% 0% Grama, tall POACEAE Bouteloua various 0% 0% 0% 100% 0% Buffalograss POACEAE Buchloe various 0% 0% 0% 100% 0% W. bumelia SAPOTACEAE Bumelia lanuginosa 100% 0% 0% 0% 0% American Beauty Berry VERBENACEAE Callicarpa americana 0% 50% 50% 0% 0% Winecup MALVACEAE Callirhoe various 0% 0% 0% 100% 0% Eveningprimrose ONAGRACEAE Calylophus various 100% 0% 0% 0% 0% Halfshrub sundrop ONAGRACEAE Calylophus various 0% 0% 0% 0% 0% Thistle ASTERACEAE Carduus austrinus 10% 90% 0% 0% 0% Pecan FABACEAE Carya illinoesis 100% 0% 0% 0% 0% Coffee senna FABACEAE Cassia fasciculata 100% 0% 0% 0% 0% Showy partridge pea FABACEAE Cassia occidentalis 100% 0% 0% 0% 0% Scarlet Paint Brush SCROPHULARIACEAE Castilleja Englemann 0% 100% 0% 0% 0% Hackberry ULMACEAE Celtis occidentalis 25% 0% 0% 0% 75% Sugar hackberry ULMACEAE Celtis laevigata 25% 0% 0% 0% 75% Grass Bar POACEAE 0% 50% 0% Cenchrus incertus 50% 0% Basket flower ASTERACEAE Centaurea americana 0% 0% 100% 0% 0% Redbud FABACEAE Cercis canadensis 25% 25% 0% 0% 50% Prairie senna FABACEAE 100% Chamaecrista fasciculata 0% 0% 0% 0% Partridge Pea FABACEAE Chamaecrista Michaux 0% 0% 100% 0% 0% Windmillgrass POACEAE Chloris various 50% 0% 0% 50% 0% Yellow spine thistle ASTERACEAE Cirsum ochrocentrum 50% 50% 0% 0% 0% Wavyleaf thistle ASTERACEAE Cirsum undulatum 50% 50% 0% 0% 0% Virginia springbeauty PORTULACCACEAE Claytonia virginica 100% 0% 0% 0% 0% Bullnettle EUPHORBIACEAE 0% Cnidoscolus texanus 100% 0% 0% 0% Carolina jointtail 0% 0% POACEAE Coelorachis various 0% 0% 100% Dayflower COMMELINACEAE 0% 0% Commelina various 100% 0% 0% Hawthorn ROSACEAE 0% Crataegus various 100% 0% 0% 0% Haw ROSACEAE sp. various 100% 0% 0% 0% 0% Croton EUPHORBIACEAE Croton various 0% 100% 0% 0% 0% Yellow Nutsedge CYPERACEAE Cyperus esculentus 100% 0% 0% 0% 0% Dalea FABACEAE Dalea various 100% 0% 0% 0% 0% Prairie-clover FABACEAE Dalea purpurea 100% 0% 0% 0% 0% White prairie clover FABACEAE Dalea candide 75% 25% 100% 0% 0%
286
Table A10. Economic values: Vegetation scores for soil settings.
Common Name Family Genus Species Edibility Medicinal Toxicity Forage Other
Bundleflower FABACEAE Desmanthus various 0% 0% 0% 100% 0%
Tickclover FABACEAE Desmodium various 0% 0% 0% 100% 0%
Arizona cottontop POACEAE Digitaria californica 50% 0% 0% 50% 0%
Fall witchgrass various POACEAE Digitaria 0% 0% 0% 100% 0%
Wood Sorrel OXALIDACEAE Jacquin 0% dillenii 0% 100% 0% 0%
Common persimmon EBENACEAE virginiana Diospyros 100% 0% 0% 0% 0%
Persimmon Diospyros virginiana L. 0% EBENACEAE 50% 50% 0% 0%
Blacksamson ASTERACEAE Echinacea angustifolia 0% 100% 0% 0% 0%
Coneflower ASTERACEAE Echinacea various 0% 100% 0% 0% 0%
Wildrye, Canada POACEAE Elymus various 50% 0% 0% 50% 0%
Wildrye, Virginia POACEAE Elymus various 0% 50% 0% 0% 50%
Engelmann-daisy ASTERACEAE Engelmannia various 100% 0% 0% 0% 0%
Ephedra EPHEDRACEAE Ephedra various 0% 0% 100% 0% 0%
Lovegrass, plains POACEAE Eragrostis various 0% 50% 0% 0% 50%
Lovegrass, purple POACEAE Eragrostis various 0% 50% 0% 0% 50%
Lovegrass, red POACEAE Eragrostis various 50% 0% 0% 50% 0%
Lovegrass, sand POACEAE Eragrostis various 0% 50% 50% 0% 0%
Texas cupgrass POACEAE Eriochloa various 0% 50% 0% 50% 0%
L. wild buckwheat POLYGONACEAE Eriogonum longifolium 100% 0% 0% 0% 0%
Buckwheat POLYGONACEAE Eriogunum various 100% 0% 0% 0% 0%
White ash OLEACEAE Fraxinus americana 50% 0% 0% 0% 50%
Snakecotton AMARANTHACEAE Frielichia various 0% 100% 0% 0% 0%
Downy milkpea FABACEAE Galactia volubilis 75% 25% 0% 100% 0%
Honey locust FABACEAE Gleditsia tricanthos 40% 0% 10% 0% 50%
C. sunflower ASTERACEAE Helianthus annus 0% 50% 50% 0% 0%
M. sunflower ASTERACEAE Helianthus various 50% 50% 0% 0% 0%
Sunflower ASTERACEAE Helianthus anmus l. 50% 0% 50% 0% 0%
Curlymesquite POACEAE Hilaria various 50% 0% 0% 50% 0%
Western Indigo FABACEAE Indigofera various 0% 0% 0% 100% 0%
Black Walnut FABACEAE Juglans nigra L. 100% 0% 0% 0% 0%
Rush FABACEAE Juncus various 100% 0% 0% 0% 0%
Juniper CUPRESSACEAE Juniperus various 75% 0% 0% 25% 0%
Red Cedar PINACEAE Juniperus virginiana 50% 0% 25% 0% 25%
Trailing ratany FABACEAE Krameria various 100% 0% 0% 0% 0%
G. sprangletop POACEAE Leptochloa various 0% 50% 0% 0% 50%
Lespedeza FABACEAE Lespedeza various 0% 0% 50% 0% 50%
Honeysuckle CAPRIFOLIACEAE Lonicera various 0% 0% 0% 0% 0%
T.Honeysuckle CAPRIFOLIACEAE Lonicera sempervirens 0% 100% 0% 0% 0%
Bluebonnet FABACEAE Lupinus texensis 0% 0% 0% 100% 0%
Mint FABACEAE Mentha sp. various 0% 100% 0% 0% 0%
Catclaw FABACEAE Mimosa 50% various 50% 0% 0% 0%
Sensitivebriar FABACEAE Mimosa various 0% 0% 0% 50% 50%
Horsemint LAMIACEAE 100% Monarda citriodora C. 0% 0% 0% 0%
Red mulberry MORACEAE rubra 0% Morus 75% 25% 0% 0%
Seep muhly POACEAE Muhlenbergia various 0% 50% 0% 0% 50%
T. wintergrass POACEAE Nassella various 50% 0% 0% 50% 0%
Water lily NYMPHAEACEAE Nelubo 0% 0% 0% 0% lutea 100%
287
Table A11. Economic values: Vegetation scores for soil settings. Common Name Family Genus Species Edibility Medicinal Toxicity Forage Other Common prickly pear Opuntia compressa 0% 0% 0% CACTACEAE 75% 25% Prickly Pear CACTACEAE Opuntia phaeacantha E. 50% 50% 0% 0% 0% Agrito OXALIDACEAE corniculata 0% 75% 0% 0% Oxalis 25% Panicum, low POACEAE Panicum various 50% 0% 0% 50% 0% Panicum, Scribner POACEAE Panicum various 0% 0% 50% 0% 50% Switchgrass POACEAE Panicum various 50% 0% 0% 50% 0% Vine-mesquite POACEAE Panicum various 0% 0% 0% 100% 0% Western wheatgrass POACEAE Pascopyrum 0% 0% various 50% 0% 50% Paspalum, Florida POACEAE Paspalum 50% various 50% 0% 0% 0% Paspalum, fringeleaf POACEAE Paspalum various 50% 0% 0% 50% 0% Scurfpea FABACEAE Pediomelum various 100% 0% 0% 0% 0% Penstemon SCROPHULARIACEAE Penstemon various 0% 0% 0% 0% 0% Purple groundcherry SOLANACEAE lobata 0% 0% Physalis 75% 0% 25% Common pokeberry PHYTOLACCACEAE Phytolacca various 0% 50% 0% 25% 25% Buckthorn PLANTAGINACEAE Plantago various 0% 0% 0% 0% 0% Plantain PLANTAGINACEAE Plantago sp. various 50% 50% 0% 0% 0% Texas bluegrass POACEAE Poa various 50% 0% 0% 50% 0% Milkwort POLYGALACEAE Polygala various 75% 25% 0% 0% 0% Cottonwood SALICACEAE Populus D. 0% 50% 0% 0% 50% Common mesquite FABACEAE Prosopis juliflora 70% 0% 0% 0% 30% Honey mesquite FABACEAE Prosopis glandulosa 70% 0% 0% 0% 30% Mesquite Prosopis 0% 0% 30% FABACEAE glandulosa C 70% 0% American plum ROSACEAE Prunus 100% 0% 0% 0% americana 0% Chickasaw plum ROSACEAE Prunus angustifolia 80% 10% 0% 0% 10% Common chokecherry ROSACEAE Prunus virginiana 100% 0% 0% 0% 0% Oklahoma plum ROSACEAE Prunus gracilis 100% 0% 0% 0% 0% Mexican plum ROSACEAE Prunus 10% 0% 10% mexicana 80% 0% Wild plum ROSACEAE Prunus sp. various 100% 0% 0% 0% 0% Indian breadroot FABACEAE Psoralea hypogeae 100% 0% 0% 0% 0% Wild alfafa FABACEAE Psoralidium various 50% 0% 0% 50% 0% Caroline false-dandelion ASTERACEAE Pyrrhopappus carolinianus 50% 50% 0% 0% 0% Oak, bur FABACEAE Quercus macrocarpa 100% 0% 0% 0% 0% Oak, blackjack FABACEAE Quercus marilandica 100% 0% 0% 0% 0% Oak, willow FAGACEAE Quercus phellos L. 0% 0% 0% 0% 100% Shumard oak FABACEAE Quercus shumardii 100% 0% 0% 0% 0% Oak FABACEAE Quercus sp. various 100% 0% 0% 0% 0% Oak, post FABACEAE Quercus stellata 100% 0% 0% 0% 0% Oak, live FAGACEAE Quercus virginiana Mill 0% 25% 0% 0% 75% Skink bush aromatica 0% ANACARDIACEAE Rhus 10% 75% 0% 15% Flame leaf sumac ANACARDIACEAE 0% Rhus copallina 50% 25% 0% 25% Smooth sumac ANACARDIACEAE Rhus glabra 25% 50% 25% 0% 0% Sumac 25% 0% ANACARDIACEAE Rhus spp. various 50% 0% 25% Poison Ivy ANACARDIACEAE Rhus toxicodendron L. 0% 100% 0% 0% 0% Snoutbean 0% 50% FABACEAE Rhynchosia various 50% 0% 0% Clove current SAXIFRAGACEAE 0% 0% 0% Ribes odoratum 100% 0% Black Locust FABACEAE Robinia pseudo-acacia L. 0% 0% 25% 0% 75% Black raspberry ROSACEAE Rubus occidentialis 100% 0% 0% 0% 0% Dewberry ROSACEAE Rubus aboriginum 100% 0% 0% 0% 0% Red raspberry ROSACEAE Rubus idaeaus 100% 0% 0% 0% 0% Louisiana blackberry ROSACEAE Rubus lousianus 100% 0% 0% 0% 0% Southern Dewberry ROSACEAE Rubus triviklis M. 100% 0% 0% 0% 0% Ruella ACANTHACEAE Ruellia various 0% 0% 0% 0% 100% Arrowhead ALISMATACEAE Sagittaria latiforlia 100% 0% 0% 0% 0% Black Willow SALICACEAE Salix nigra Marsh. 0% 100% 0% 0% 0% Sage FABACEAE Salvia various 100% 0% 0% 0% 0% Bluestem, little 50% 0% POACEAE Schizachyrium various 50% 0% 0% Soft stem bulrush CYPERACEAE Scirpus validus 100% 0% 0% 0% 0% Skullcap various 100% 0% 0% FABACEAE Scutellaria 0% 0% Knottroot bristlegrass POACEAE various Setaria 50% 0% 0% 50% 0% Sida MALVACEAE 100% 0% 0% 0% 0% Sida various
288
Table A12. Economic values: Vegetation scores for soil settings. Common Name Family Genus Species Edibility Medicinal Toxicity Forage Other Bumelia Sideroxylon 0% 0% 0% 0% SAPOTACEAE various 0% Compassplant ASTERACEAE Silphium laciniatum 0% 100% 0% 0% 0% Bushsunflower ASTERACEAE 0% 100% 0% 0% 0% Simsia various Greenbriar SMILACACEAE Smilax various 100% 0% 0% 0% 0% Saw greenbriar SMILACACEAE Smilax bona-nox 75% 0% 0% 0% 25% Indiangrass POACEAE Sorghastrum various 50% 0% 0% 50% 0% Dropseed, meadow POACEAE Sporobolus various 50% 0% 0% 50% 0% Dropseed, sand POACEAE Sporobolus various 50% 0% 0% 50% 0% Dropseed, tall POACEAE Sporobolus various 50% 0% 0% 50% 0% Falsegaura ONAGRACEAE various Steinosiphon 0% 0% 0% 0% 0% Queensdelight EUPHORBIACEAE Stillingia various 0% 0% 0% 0% 0% Wildbean FABACEAE Strophostyles various 80% 10% 0% 10% 0% Trailing wild bean FABACEAE Strophostyles helvola 80% 10% 0% 10% 0% Coralberry CAPRIFOLIACEAE Symphoricarpos various 0% 0% 100% 0% 0% Tridens, purpletop POACEAE Tridens various 50% 0% 0% 50% 0% Tridens, rough POACEAE Tridens various 50% 0% 0% 50% 0% Tridens, slim POACEAE Tridens various 50% 0% 0% 0% 50% Tridens, white 50% 0% 0% POACEAE Tridens various 0% 50% Clover FABACEAE Trifolium repens L. 25% 25% 0% 50% 0% Eastern gamagrass POACEAE Tripsacum various 50% 0% 0% 50% 0% Cattail TYPHACEAE Typha latifolia 90% 0% 10% 0% 0% Cedar Elm ULMACEAE Ulmus crassifolia Nutt 0% 0% 0% 0% 100% Elm, American ULMACEAE Ulmus americana L. 25% 0% 0% 0% 75% Elm, cedar ULMACEAE 0% 0% 0% 0% 100% Ulmus various Elm, winged ULMACEAE Ulmus various 0% 0% 0% 0% 100% Slippery elm ULMACEAE Ulmus rubra 25% 0% 0% 0% 75% Blue Vervain VERBENACEAE 0% 100% 0% 0% 0% Verbena halei Verbena VERBENACEAE Verbena 0% 0% various 100% 0% 0% Ironweed ASTERACEAE Verononia various 0% 0% 0% 0% 0% Haw ROSACEAE Viburnum prunifolium 0% 0% 75% 25% 0% Vetch 0% 0% 0% 0% FABACEAE Vicia various 100% Grape VITACEAE Vitis spp. various 50% 50% 0% 0% 0% Soap weed AGAVACEAE Yucca glauca 50% 50% 0% 0% 0% Yucca AGAVACEAE Yucca louisianensis T. 0% 0% 50% 25% 25% Prickly Ash RUTACEAE Zanthoxylum clavaherculis L. 0% 100% 0% 0% 0% Indian Corn FABACEAE 100% 0% 0% 0% 0% Zea mays various Lotebush RHAMNACEAE Ziziphus various 100% 0% 0% 0% 0% Dropseed, hairy POACEAE various 50% 0% 0% 0% 50% Grain Sorghum POACEAE 50% 0% 0% 50% various 0% L. hackberry ULMACEAE various 0% 0% 0% 0% 100% Oak, shin FAGACEAE various 0% 0% 100% 0% 0% Oak, Texas FAGACEAE various 0% 50% 50% 0% 0% Peanuts FABACEAE various 100% 0% 0% 0% 0% Roughleaf dogwood CORNACEAE 0% 0% 0% 100% various 0% Sedge CYPERACEAE 100% 0% 0% 0% various 0% Stemless actinea POACEAE various 50% 0% 50% 0% 0% Texas ash OLEACEAE various 0% 0% 0% 0% 100% Texas sophora FABACEAE various 0% 0% 0% 0% 0% Western soapberry SAPINDACEAE 0% 0% 100% 0% 0% various Wheat POACEAE various 100% 0% 0% 0% 0%
289
Table A13. Economic values of site catchment areas in the Upper Trinity Basin.
ECONOMIC VALUE OF TRINITY RIVER WEST FORK AND ELM FORK SITE CATCHMENTS BY OCCUPATION FREQUENCY ARCHAEOLOGICAL SITES EDIBILITY VALUE MEDICINAL VALUE SUPPLEMENTAL VALUE ECONOMIC VALUE TRINITY WEST FORK Sum Mean St Dev Var/Mean Sum Mean St Dev Var/Mean Sum Mean St Dev Sum Var/Mean Mean St Dev Var/MeanARCHAIC 0.1800 0.0450 0.0107 0.0026 0.1576 0.0394 0.0116 0.0034 0.1709 0.0427 0.0074 0.0013 0.1689 0.0422 0.0095 0.00214EARLY ARCHAIC 0.1324 0.0441 0.0122 0.1257 0.0419 0.0073 0.0013 0.0034 0.1176 0.0392 0.0133 0.0045 0.1242 0.0414 0.0107 0.00276MIDDLE ARCHAIC 0.1902 0.0475 0.0121 0.0081 0.0031 0.1743 0.0436 0.0138 0.0044 0.1790 0.0448 0.0014 0.1788 0.0447 0.0109 0.00265 LATE ARCHAIC 0.0108 0.0025 0.2133 0.0427 0.0125 0. 0.0443 0.0098 0.00217 0.2360 0.0472 0037 0.2196 0.0439 0.0740 0.1247 0.2213LATE PREHISTORIC 0.1596 0.0399 0.0129 0.1541 0.1138 0.34492 0.0041 0.1397 0.0349 0.0137 0.0054 0.0385 0.0087 0.0019 0.1503 0.0376 LATE PREHISTORIC I 0.0568 0.0568 0.0000 0.0000 0.0479 0.0000 0.0551 0.0551 0.0000 0.0479 0.0000 0.0000 0.0524 0.0524 0.0000 0.0000 LATE PREHISTORIC II 0.1543 0.0514 0.0039 0.1450 0.0019 0.0001 0.1444 0.0481 0.0032 0.0003 0.1360 0.0453 0.0069 0.0011 0.0483 0.00021 PREHISTORIC 0.5838 0.0486 0.0081 0. 0.0089 0.0017 0.5498 0.0458 0.0078 0014 0.5128 0.0427 0.0101 0.0024 0.5634 0.0469 0.00133 TRINITY ELM FORK ARCHAIC 0.0120 0.0033 1.1491 0.0522 0.0187 0.0067 1.2124 0.0551 0.0135 0.0033 0.9599 0.0436 1.1196 0.0509 0.0137 0.00369EARLY ARCHAIC 0.7376 0.0461 0.0128 0.0036 0.8137 0.0509 0.0039 0.0003 0.6638 0.0414 0.0047 0.0005 0.7446 0.0465 0.0045 0.00044MIDDLE ARCHAIC 0.0018 0.2593 0.0432 0.0059 0.0008 0.0444 0.0120 0.2879 0.0479 0.0001 0.0046 0.0005 0.2522 0.0420 0.2665 0.00324 LATE ARCHAIC 0.0026 1.2129 0.0433 0.0118 1.5218 0.0544 0.0120 0.0032 1.3979 0.0499 0.0183 0.0067 1.3971 0.0499 0.0123 0.00303LATE PREHISTORIC 1.6021 0.0517 0.0050 0.0005 1.3266 0.0428 0.0056 0.0007 1.4584 0.0470 0.0139 0.0041 1.4785 0.0477 0.0061 0.00078 LATE PREHISTORIC I 0.5468 0.0497 0.0032 0.4859 0.0002 0.4871 0.0443 0.0059 0.0008 0.0442 0.0110 0.0028 0.5078 0.0462 0.0046 0.00046 LATE PREHISTORIC II 0.6994 0.0119 0.0500 0.0028 0.0002 0.5850 0.0418 0.0045 0.0005 0.6075 0.0434 0.0033 0.6396 0.0457 0.0044 0.00042 PREHISTORIC 2.5729 0.0547 0.0091 0.0015 1.9450 0.0414 0.0098 0.0023 2.2058 0.0469 0.0147 0.0046 2.3181 0.0493 0.0081 0.00133
ECONOMIC VALUE OF TRINITY RIVER WEST FORK AND ELM FORK SITE CATCHMENTS BY OCCUPATION FREQUENCY ARCHAEOLOGICAL SITES EDIBILITY VALUE MEDICINAL VALUE SUPPLEMENTAL VALUE ECONOMIC VALUE TRINITY WEST FORK Sum Mean St Dev Var/Mean Sum Mean St Dev Var/Mean Sum Mean St Dev Var/Mean Sum Mean St Dev Var/MeanSINGLE 0.5800 0 0.0486 0.0080 0.0013 0.512 0.0420 0.0100 0.0024 0.5800 0.0486 0.0080 0.0013 0.5490 0.0450 0.0080 0.00142DUAL 0.0079 0.1532 0.0511 0.0049 0.0005 0.1367 0.0456 0.0014 0.1448 0.0483 0.0055 0.0006 0.1442 0.0481 0.0050 0.00052TRI 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000MULTI 0.1324 0.0441 0.0122 0.0034 0.1176 0.0392 0.0133 0.0045 0.1257 0.0419 0.0073 0.0013 0.1242 0.0414 0.0107 0.00277 TRINITY ELM FORK SINGLE 3.1760 0.0530 0.0080 0.0012 2.4400 0.0410 0.0080 0.0016 2.8000 0.0470 0.0140 0.0042 2.8800 0.0480 0.0070 0.00102DUAL 0.9838 0.0578 0.0143 0.0035 0.7718 0.0454 0.0141 0.0044 0.9115 0.0536 0.0227 0.0096 0.9043 0.0532 0.0150 0.00423TRI 0.4098 0.0512 0.0042 0.0003 0.3451 0.0431 0.0062 0.0009 0.3909 0.0489 0.0063 0.0008 0.3270 0.0478 0.0036 0.00027MULTI 0.4500 0.0500 0.0024 0.0001 0.3772 0.0419 0.0052 0.0006 0.3889 0.0432 0.0115 0.0031 0.4110 0.0457 0.0044 0.00042
290
Table A14. Vegetation potentials of site catchment areas in the Upper Trinity Basin.
POTENTIAL VEGETATION TYPE WITHIN TRINITY RIVER WEST FORK AND ELM FORK SITE CATCHMENTS BY OCCUPATION FREQUENCY ARCHAEOLOGICAL SITES GRAINS AND SEED CROPS GRASSES AND LEGUMES WILD HERBACEOUS PLANTS SHRUBS
TRINITY WEST FORK Sum Mean St Dev Var/Mean Sum Mean
St Dev Var/Mean Sum Mean
St Dev Var/Mean Sum Mean
St Dev Var/Mean
ARCHAIC 0.918 0.0720.230 0.0224 1.307 0.327 0.129 0.0505 1.551 0.388 0.117 0.0350 1.688 0.422 0.117 0.03237EARLY ARCHAIC 0.678 0.226 0.077 0.0260 0.913 0.304 0.141 0.0657 1.033 0.344 0.107 0.0331 1.133 0.378 0.105 0.02933MIDDLE ARCHAIC 1.014 0.254 0.082 0.0264 1.277 0.319 0.125 0.0491 1.360 0.340 0.093 0.0253 1.546 0.386 0.092 0.02205LATE ARCHAIC 1.233 0.247 0.074 0.0225 1.667 0.333 0.116 0.0400 1.853 0.371 0.103 0.0287 2.076 0.415 0.101 0.02438LATE PREHISTORIC 0.781 0.195 0.085 0.115 0.0437 1.0.0374 1.119 0.280 0.130 0.0601 1.218 0.304 326 0.332 0.121 0.04439 LATE PREHISTORIC I 0.276 0.276 0.000 0.0000 0.408 0.408 0.000 0.0000 0.397 0.397 0.000 0.0000 0.410 0.410 0.000 0.000LATE PREHISTORIC II 0.866 0.289 0.013 0.0005 1.213 0.404 0.003 0.0000 1.303 0.434 0.029 0.0019 1.410 0.470 0.044 0.00403PREHISTORIC 0.407 3.248 0.271 0.075 0.0206 4.584 0.324 0.104 0.0331 4.882 0.096 0.0224 5.231 0.436 0.076 0.01332 TRINITY ELM FORK ARCHAIC 6.093 0.277 0.086 0.0266 0.322 7.092 0.096 0.0286 8.352 0.380 0.104 0.0284 9.094 0.413 0.130 0.04089EARLY ARCHAIC 4.581 0.286 0.094 0.0306 5.263 0.329 0.095 0.0274 6.000 0.375 0.101 0.0271 5.919 0.370 0.107 0.03078MIDDLE ARCHAIC 1.803 0.300 0.063 0.0130 1.953 0.325 0.064 0.0124 2.180 0.363 0.059 0.0096 2.008 0.334 0.058 0.01007LATE ARCHAIC 7.324 0.264 0.091 0.0316 8.454 0.389 0.302 0.099 0.0325 10.337 0.369 0.109 0.0321 10.887 0.133 0.04543 LATE PREHISTORIC 7.483 0.241 0.084 0.02519 0.0290 8.539 0.275 0.083 0.0253 10.730 0.346 0.075 0.0163 11.061 0.357 0.095LATE PREHISTORIC I 2.966 0.270 0.772 2.2102 3.289 0.299 0.068 0.0154 3.818 0.347 0.072 0.0149 3.783 0.344 0.082 0.01931LATE PREHISTORIC II 0.087 3.548 0.253 0.0299 4.028 0.288 0.092 0.0293 4.525 0.347 0.088 0.0224 4.937 0.353 0.101 0.02879PREHISTORIC 1 0.0279 0.102 11.901 0.253 0.102 0.0409 3.721 0.292 0.122 0.0509 14.876 0.317 0.094 16.133 0.343 0.03049
POTENTIAL VEGETATION TYPE WITHIN TRINITY RIVER WEST FORK AND ELM FORK SITE CATCHMENTS BY OCCUPATION FREQUENCY ARCHAEOLOGICAL SITES GRAINS AND SEED CROPS GRASSES AND LEGUMES WILD HERBACEOUS SHRUBS
TRINITY WEST FORK Mean Dev SumSt
Var/Mean Sum MeanSt Dev Var/Mean Sum Mean
St Dev Var/Mean Sum Mean
St Dev Var/Mean
SINGLE 4.689 0.0820.390 0.0172 0.269 3.232 0.077 0.0221 4.872 0.406 0.097 0.0232 5.188 0.432 0.085 0.01672 DUAL 0.00558 1.159 0.0170.386 0.0007 0.861 0.287 0.049 0.0084 1.286 0.429 0.073 0.0124 1.455 0.485 0.052TRI 0.000 0.000 0.000 0.000 0.000 0.0000.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000MULTI 0.913 0.304 0.141 0 0.344 0.107 0.0331 1. .0657 678 0.226 0. 0.077 0.0260 1.033 133 0.378 0.105 0.02929 TRINITY ELM FORK SINGLE 16.888 0.286 0.105 0.251 0.104 0.0385 14.838 0.096 0.0367 19.522 0.330 0.098 0.0291 20.835 0.353 0.03064DUAL 0.272 0.163 5.571 0.1400.327 0.0599 4.626 0.113 0.0469 6.234 0.366 0.133 0.0483 6.741 0.396 0.06709
TRI 0.0055 2.416 0.0660.302 0.0144 2.228 0.279 0.070 0.0176 2.943 0.368 0.045 3.138 0.392 0.061 0.00949MU 3. 0.0190 LTI 2.633 0.293 0.073 0.0182 0.255 2.300 0.079 0.0245 107 0.345 0.081 3.107 0.345 0.088 0.02245
291
Table A15. Wildlife potentials of site catchment areas in the Upper Trinity Basin.
WILDLIFE POTENTIAL WITHIN TRINITY RIVER WEST FORK AND EAST FORK SITE CATCHMENTS BY TIME PERIOD ARCHAEOLOGICAL SITES OPENLAND WILDLIFE RANGELAND WILDLIFE WETLAND WILDLIFE WILDLIFE
TRINITY WEST FORK Sum Mean St Dev Var/Mean Sum Mean
St Dev Var/Mean Sum Mean
St Dev Var/Mean Sum Mean
St Dev Var/Mean
ARCHAIC 0.333 0.0378 1.332 0.112 1.483 0.371 0.105 0.0298 0.110 0.028 0.007 0.0019 2.926 0.732 0.224 0.06856 EARLY ARCHAIC 1.034 0.344 0.108 0.679 0.924 0.308 0.119 0.0458 0.0339 0.078 0.026 0.008 0.0022 2.036 0.234 0.08069MIDDLE ARCHAIC 0.095 1.307 0.325 0.107 0.0353 1.412 0.353 0.0257 0.116 0.029 0.008 0.0025 2.827 0.707 0.209 0.06159 LATE ARCHAIC 1.691 0.332 0.099 0.0297 0.05193 1.843 0.369 0.091 0.0224 0.146 0.029 0.076 0.1968 3.680 0.736 0.196 LATE PREHISTORIC 0.0446 0.241 1.074 0.269 0.124 0.0568 1.220 0.305 0.117 0.119 0.030 0.009 0.0030 2.413 0.603 0.09645LATE PREHISTORIC I 0.000 0.0000 0.032 0.380 0.380 0.000 0.0000 0.383 0.383 0.032 0.000 0.0000 0.800 0.800 0.000 0.000LATE PREHISTORIC II 0.095 0.0000 1.205 0.402 0.017 0.0008 1.307 0.436 0.034 0.0027 0.032 0.001 2.607 0.869 0.051 0.00299PREHISTORIC 4.290 0.357 0.091 0.0017 9.515 0.793 0.0231 4.841 0.403 0.086 0.0182 0.385 0.032 0.007 0.172 0.03709 TRINITY ELM FORK ARCHAIC 7.099 0.323 0.079 0.0192 8.007 0.364 0.105 0.0304 1.385 0.063 0.031 0.0150 0.179 16.491 0.750 0.04274EARLY ARCHAIC 5.233 0.327 0.094 0.0272 5.642 0.353 0.098 0.0272 1.106 0.069 0.033 0.0158 11.982 0.748 0.200 0.05332 MIDDLE ARCHAIC 0.0117 0.324 0.053 0.0086 0.536 0.089 0.031 0.0108 4.385 0.738 0.123 0.02064 1.907 0.318 0.061 1.942LATE ARCHAIC 8.593 0.307 0.089 0.0259 9.783 0.349 0.108 0.072 0.728 0.05644 0.0333 2.015 0.035 0.0167 20.391 0.203LATE PREHISTORIC 8.658 0.279 0.073 0.077 0.0212 10.058 0.324 0.0162 2.386 0.077 0.049 0.0307 21.103 0.681 0.147 0.03175LATE PREHISTORIC I 3.270 0.297 0.068 0.0155 3.490 0.317 0.060 0.0113 0.919 0.084 0.041 0.0204 7.679 0.698 0.151 0.0327LATE PREHISTORIC II 4.088 0.292 0.090 0.0279 4.517 0.323 0.078 0.0189 0.949 0.068 0.031 0.0141 9.553 0.682 0.185 0.05001PREHISTORIC 13.418 0.285 0.100 0.0351 14.419 0.307 0.084 0.0229 3.464 0.074 0.045 0.0274 31.301 0.665 0.191 0.05457
WILDLIFE POTENTIAL WITHIN TRINITY RIVER WEST FORK AND EAST FORK SITE CATCHMENTS BY TIME PERIOD ARCHAEOLOGICAL SITES OPENLAND WILDLIFE RANGELAND WILDLIFE WETLAND WILDLIFE WILDLIFE VALUE
TRINITY WEST FORK Sum Sum Mean St Dev Var/Mean Mean
St Dev Var/Mean Sum Mean
St Dev Var/Mean Mean
St Dev Sum Var/Mean
SINGLE 4.299 9. 0.358 0.089 0.0221 4.830 0.402 0.087 0.0188 0.411 0.034 0.005 0.0007 541 0.795 0.165 0.03425DUAL 1.189 0.396 0.019 0.0009 1.277 0.426 0.037 0.0032 0.101 0.034 0.003 0.0003 2.566 0.855 0.053 0.00329 TRI 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000MULTI 0.0458 034 0.345 0.109 0.0343 0.0.924 0.308 0.119 1. 078 0.026 0.008 0.0022 2.036 0.679 0.234 0.08083 TRINITY ELM FORK SINGLE 16.871 .685 0.285 0.093 0.0303 18.998 0.322 0.090 0.0252 4.569 0.077 0.053 0.0365 40.440 0 0.182 0.04836DUAL 5.451 0.321 0.121 0.0456 6.059 0.356 0.135 0.0512 1.119 0.065 0.030 0.0138 12.631 0.743 0.254 0.08683TRI 2.506 0.313 0.050 0.0080 2.729 0.341 0.044 0.0057 0.645 0.081 0.035 0.0151 5.884 0.735 0.101 0.01388MULTI 2.626 0.292 0.074 0.0188 2.876 0.319 0.067 0.0141 0.619 0.068 0.036 0.0191 6.119 0.679 0.159 0.03723
292
GIS GLOSSARY
Buffer: Procedure conducted with ArcMap Geoprocessing tools which generates a circle or polygon at a specified distance around a point.
Dissolve: Procedure conducted with ArcMap Geoprocessing tools that “creates a new data set in which all features in an input layer that have the same value for a specified attribute become a single feature” (Ormsby et al., 2001: 270).
Overlay analysis: A method of analysis that “creates a new data set by superimposing one layer on another. The output data has the attributes of both input layers” (Ormsby et al., 2001: 306).
Raster: An image that uses a “grid structure to store geographic information” (Minami, 2000: 510).
Selection: Selecting data can be conducted in ArcMap by location or attributes. Selecting features by location involves specifying “a selection method, a selection layer, a spatial relationship, a reference layer, and sometimes a distance buffer” (Ormsby et al., 2001: 250). Selecting features by attribute involves specifying a specific attribute (i.e. time period) to create a new data layer. Symbology: A term that refers to “criteria used to determine symbols for the features in a layer” (Minami, 2000: 513). Tabular data: A phrase that refers to “descriptive information that is stored in rows and colums and can be linked to map features” (Minami, 2000: 513).
Clip: Procedure conducted with ArcMap Geoprocessing tools that “trims features in one layer at the boundaries of features in another layer” (Ormsby et al., 2001: 282).
Joining: To associate a non-spatial table with a spatial data layer based on a unique identification code (i.e. soils series). This assigns meaning to a spatial data layer.
Projection: the transformation of descriptive data to spatial location with a geographic coordinate system. Query: A procedure conducted in ArcMap to extract data that involves “a question or request for selecting features. A query often appears in the form of a statement or logical expression. In ArcMap, a query contains a field, an operator, and a value” (Minami, 2000: 510).
293
REFERENCES
Urbanism. edited by Peter Ucko, Ruth Tringham and G.W. Dimbleby. Proceedings
Allen, K.M.S., Green, S.W. and Zubrow, E.B.W. Interpreting Space: GIS and Archaeology,
.
Biesaart, L.A., W.R. Roberson, L.C. Spotts. Prehistoric Archaeological Sites in Texas: A Statistical Overview. Office of the State Archaeologist Special Report 28. Austin: Texas Historical Commission, 1985.
Binford, L.R. “The Archaeology of Place.” Journal of Anthropological Archaeology 1(1)
(1982): 5-31. _____.Constructing Frames of Reference: An Analytical Method for Archaeological
Theory Building Using Hunter-Gatherer and Environmental Data Sets. Berkeley, California; London : UC Press, 2002.
Bousman, B. and L. Verrett. An Archaeological Reconnaissance at Aubrey Reservoir.
Archaeology Research Program, Southern Methodist University, Dallas. Fort Worth: U.S. Army Corps of Engineers, 1973.
Bruseth, James E., Daniel E. McGregor, and William A. Martin. “Hunter-Gatherers of the
Allan, W. “Ecology, Techniques and Settlement Patterns.” In: Man, Settlement and
of a meeting of the Research Seminar in Archaeology and Related Subjects held at the Institute of Archaeology, London University. Cambridge, Massachusetts: Schenkman Publishing Company, 1973.
London: Taylor and Francis, 1990.
Bernick, Kathryn N. A Site Catchment Analysis of the Little Qualicum River Site, DiSc 1, Wet Site on the East Coast of Vancouver Island, B C. Ottawa: National Museums of Canada, 1983.
Brown, K.L. and B.C. Yates. Archaeological Testing of the Lewisville Lake Shoreline, Denton County, Texas. Denton: Institute of Applied Sciences, University of North Texas, 1990.
Brown, K.L. and S.A. Lebo. Archaeological Testing of the Lewisville Lake Shoreline,
Denton County, Texas. Denton: Institute of Applied Sciences, University of North Texas, 1991.
Prairie Margin: Summary of the Prehistoric Archaeological Record.” In: Hunter-Gatherer Adaptations Along the Prairie Margins: Site Excavations and Synthesis of Prehistoric Archaeology, edited by D.E. McGregor and J.E. Bruseth. Dallas: Richland Creek Technical Series, Volume III. Archaeology Research Program, Southern Methodist University, 1987.
294
Butzer, Karl M. Environment and Archeology: An Ecological Approach to Prehistory. Chicago: Aldine-Atherton, 1971.
Christaller, Walter. Central Places in Southern Germany. Translated from Die zentralen
Orte in Süddeutschland by Carlisle W. Baskin. New Jersey: Englewood Cliffs, Prentice-Hall, 1966.
f
Dunnell, Robert C. Systematics in Prehistory. New York: Free Press, 1971. Dyksterhuis, E.J. “The Vegetation of the Fort Worth Prairie.” In: Ecological Monographs
_____.Archaeology as Human Ecology. Cambridge University Press: Cambridge, 1982. Chisholm, Michael. Rural Settlement and Land Use. London: Hutchinson, 1968.
Clarke, David. Models in Archaeology. London: Methuen, 1972. Coe M.D. and K.V. Flannery “Microenvironments and Mesoamerican Prehistory”, In:
Science: New Roads to Yesterday, edited by J.R. Caldwell, New York/London, 1966.
Colinvaux, Paul. Ecology 2. New York: John Wiley & Sons, Inc., 1993. Dawson, G.L. and T.L. Sullivan. Excavations at Lake Lavon: 1969. Dallas: Report
submitted to the National Park Service by the Archaeology Research Program, Department of Anthropology, Southern Methodist University, 1969.
Dennell, R.W. “The Economic Importance of Plant Resources Represented on
Archaeological Sites.” Journal o Archaeological Science 3 (1976): 229-47. Diggs, G.M., B.L. Lipscomb and R.J. O’Kennon. Shinners and Mahler’s Illustrated Flora
of North-Central Texas. Fort Worth: Botanical Research Institute of Texas, 1999. Dillehay, T.D. “Late Quaternary Bison Population Changes on the Southern Plains.”
Plains Anthropologist 19 (65) (1974): 180-196. Duke, James. Eastern/Central Medicinal Plants and Herbs. New York: Houghton Mifflin
Company, 2000.
Durham: Duke University Press, 1946. _____.“The Vegetation of the Western Cross Timbers.” In: Ecological Monographs
Durham: Duke University Press, 1948. Ferring, C.R. The Archaeology and Paleoecology of the Aubrey Clovis Site (41DN479)
Denton County, Texas. Denton: Center for Environmental Archaeology, Department of Geography, University of North Texas, 2001.
295
Ferring and B.C. Yates. An Assessment of the Cultural Resources in the Trinity Basin, Dallas, Tarrant, and Denton Counties, Texas. Denton: Institute of Applied Sciences, University of North Texas, 1986.
_____.Holocene Geoarchaeology and Prehistory of the Ray Roberts Lake Area, North- Central Texas. Denton: Institute of Applied Sciences, University of North Texas, 1997.
_____.Prehistory of Archaeological Sites at Lewisville Lake, North-Central Texas.
and Lateral Watershed, Wise County, Texas. Denton: Institute Environmental
Denton: Institute of Applied Sciences, University of North Texas, 1998 Ferring, C. Reid, Britt Bousman, Barbara Butler, Dan Crouch, Steven Hackenberger,
Steve A. Hall, Roger Hoestenbach, Kevin Leehan, Jim Robertson, and Richard Vernon. An Archaeological Reconnaissance of Fort Sill, Oklahoma. Lawton, Oklahoma: Contributions of the Museum of the Great Plains Number 6, 1978.
Filson, Roger E., and Olin F. McCormick. Archaeological Survey of Portions of Salt Creek
Studies. North Texas State University, 1975. Flannery, Kent V. "The Village and its Catchment Area: Introduction.” In: The Early
Mesoamerican village. edited by K.V. Flannery. New York: Academic Press, 1968.
_____."Empirical determination of Site Catchments in Oazaca and Tehuacan. In: The Early Mesoamerican village. edited by K.V. Flannery. New York: Academic Press, 1976.
Ford, A. and E. Pauls. Soil Survey of Denton County, Texas. : Washington D.C., U.S.
Department of Agriculture, Soil Conservation Service, U.S. Government Printing Office, 1980.
Geier, Clarence. “Some Ecological Concepts and Their Implications for Settlement
Archaeology.” Plains Anthropologist 19(63) (1974): 46-54. Goodchild, Michael F. “Geographic Information Systems and Spatial Analysis in the
Social Sciences” In: Anthropology, Space, and Geographic Information Systems. edited by Mark Aldenderfer and Herbert D.G. Maschner. Oxford University Press: New York , 1996.
Griffiths, J. and J. Bryan. The Climates of Texas Counties. College Station: Natural
Fibers Information Center, the University of Texas at Austin in Cooperation with Office of the State Climatologist, Texas A & M University: 1987.
Haggett, Peter. Locational Analysis in Human Geography. New York: St. Martin's Press,
1966.
296
Hanson, Arthur and Frankie F. Wheeler. Soil Survey, Collin County,Texas. U.S. Department of Agriculture, Soil Conservation Service, U.S. Government Printing Office: Washington D.C., 1969.
Hastorf, Christine. “Changing Resource Use in Subsistence Agricultural Groups of the
t
Higgs, E.S. and C. Vita-Finzi. “Prehistoric Economies: A Territorial Approach.” In: Papers
t
rby David L Clarke. London :Methuen, 1972.
_____.“Prehistoric economic development in sub-alpine Italy.” In: Problems in Economic and Social Archaeology edited by G. de G. Sieveking, I. H. Longworth, and K. E. Wilson. Boulder, Westview Press, 1976.
New York: Academic Press, 1976.
r t
Prehistoric Mimbres River Valley, New Mexico.” In: Modeling Change in Prehistoric Subsistence Economies, edited by Timothy K. Earle, Andrew L. Christenson. New York: Academic Press, 1980.
Hays, T, T.R. Hellier, T.E. Kennerly. Environmental Impact Study of the Elm Fork Region
of the Trini y. Arlington; Department of Biology, University of Texas at Arlington, 1972.
in Economic Prehistory : Studies by members and associates of the British Academy Major Research Project in he Early History of Agriculture; edited by E.S. Higgs. Cambridge: Cambridge University Press, 1972.
Hunt, Eleazer D. “Upgrading Site-Catchment Analyses with the use of GIS: Investigating
the Settlement Patterns of Horticulturalists.” World Archaeology. Vol. 24 (1982): 283-309.
Jarman, M.R. “A Territorial Model for Archaeology.” In: Models in A chaeology; edited
Jarman, M.R., C. Vita-Finzi and E.S. Higgs. “Site Catchment Analysis in Archaeology.”
In: Man, Settlement, and Urbanism. edited by Peter J. Ucko, Ruth Tringham, and G. W. Dimbleby. Cambridge, Mass; Morristown, N.J.: Schenkman Pub. Co.; distributed by General Learning Press, 1972.
Jochim, Michael A. Hunter-Gatherer Subsistence and Settlement: A Predictive Model.
_____.“Breaking Down the System: Recent Ecological Approaches in Archaeology.” In: Advances in Archaeological Method Theory 2. edited by M.B. Schiffer. New York: Academic Press, 1979.
_____.Strategies for Survival: Cultu al Behavior in an Ecological Con ext. New York: Academic Press, 1981.
Johnson, Eileen, and Vance T. Holliday. “The Archaic Record at Lubbock Lake.” In:
Plains Anthropologist 31 (1986): 7-54.
297
Kvamme, K.L. “Geographic Information Systems in Regional Archaeological Research and Data Management.” In: Archaeological Method and Theory 1, Tucson: The
Lynott, M.J. Archaeological Excavations at Lake Lavon 1974. Dallas: Archaeological
Research Program. Contributions in Anthropology No. 16. Southern Methodist University, 1975.
_____.“A Model of Prehistoric Adaptation in Northern Texas.” Plains Anthropologist 26 (92) (1981): 97-110.
Malczewski, Jacek. GIS and Multicriteria Decision Analysis. New York: John Wiley, 1999. McCormick, O.F., R. Filson and J. Darden. The Xerox Lewisville Archaeological Project.
McMahan, C.A., R.G. Frye and K.L. Brown. The Vegetation Types of Texas, Including
Cropland: An illustrated Synopsis to Accompany the Map. Austin: Wildlife Division, 1984.
Research Institute, Inc., 2001.
f
tTexas. Nacogdoches: Stephen F. Austin University, 1974.
University of Arizona Press, 1989. Lebo, S.A. and K.L. Brown. Archaeological Survey of the Lewisville Lake Shoreline,
Denton County, Texas. Denton: Institute of Applied Sciences, University of North Texas, 1990.
Lee, R.B. and Devore, I. Man the Hunter. Chicago: Aldine, 1968.
_____.A Regional Model for Archaeological Research in North-Central Texas. Unpublished Ph.D. Dissertation. Dallas: Southern Methodist University, 1977.
The Institute for Environmental Studies, Denton; North Texas State University, 1975.
McGregor, D.E. and J.E. Bruseth. Hunter Gatherer Adaptations Along the Prairie Margin. Dallas: Richland Creek Technical Series, Vol. III. Archaeology Research Program, Southern Methodist University, 1987.
Minami, Michael. Using ArcMap. : Redlands: ESRI Press, Environmental Systems
Newcomb, W.W., Jr. The Indians o Texas. Austin: University of Texas Press, 1986.
Nixon, Elray S. and R. Larry Willett. Vegetative Analysis of the Floodplain of the Trini y,
Nunley, Parker. An Assessment of Archaeological Resources in the Vicinity of Garza-
Little Elm Reservoir, Richland/Archaeological Society, Dallas: Miscellaneous Papers No. 1, Richland College, 1973.
298
Prewitt, Elton R. Cultural Chronology in Cen ral Texas. Bulletin of the Texas t
Ormsby, Tim, Eileen Napoleon, Robert Burke, Carolyn Groessl and Laura Feaster
Prikryl, Daniel J. A Synthesis of the Prehistory of the Lower Elm Fork of the Trinity. Unpublished Masters Thesis. Austin: University of Texas at Austin, 1987.
_____.Lower Elm Fork Prehistory : a Redefinition of Cultural Concepts and Chronologies along the Trinity, North-Central Texas. Austin: Texas Historical Commission, 1990. Renfrew, Colin. “Book Review: Peter Haggett: Locational Analysis in Human
Geography.” Antiquity 43 (1969): 74.
Press, 1977.
Department of Agriculture, Soil Conservation Service, U.S. Government Printing
Schiffer, M.B. “An Alternative to Morse’s Dalton Settlement Pattern Hypothesis.” Plains
Anthropologist 20 (1975):253-266. _____.Behavioral Archaeology. New York: Academic Press, 1976.
Sciscenti, J.V. Environmental and Cultural Resources within the Trinity Basin. Dallas:
Skinner, S.A. and L. Baird. The Archaeology and History of Lake Ray Roberts, Vol. 1: Cultural Resources Survey. Dallas: Environmental Consultants Inc., 1982a. _____.Archaeology and History of Lake Ray Roberts, Vol 2, Construction Area Testing.
Dallas: Environmental Consultants, Inc., 1982b. _____.The Archaeology and History of Lake Ray Roberts, Vol. 3: Settlement in a
Archaeological Society 52 (1981):65-89.
Getting to Know ARCGIS Desktop, Redlands: ESRI Press: Environmental Systems Research Institute, 2001.
_____.“Alternative Models for Exchange and Spatial Distribution.” In: Exchange Systems in Prehistory, edited by T.K. Earle and J.E. Ericson. New York: Academic
Ressel, Dennis D. Soil Survey of Wise County, Texas. Washington D.C.: U.S.
Office, 1989. Roper, D. "The Method and Theory of Site Catchment Analysis: A Review," In:
Advances in Archaeological Method and Theory 2, edited by M.B. Schiffer. New York: Academic Press, Inc, 1979.
SMU, Institute for the Study of Earth and Man, 1971.
Marginal Zone. Dallas: AR Consultants, Inc., 1985. _____.The Hackberry Creek Archaeological Site. Irving: AR Consultants Inc., 1999.
299
Steward, J.H. “Basin-Plateau Aboriginal Socio-Political Groups.” Bureau of Ame ican r
Urbana, University of Illinois Press, 1955.
Story, D.A. The Archaeology and Bioarchaeology of the Gulf Coastal Plain: Volume I.
Arkansas Archaeological Survey Research Series No. 38., 1990a.
Suhm, D.A. and E.B. Jelks. Handbook of Texas Archaeology: Type Descriptions. Austin:
The Texas Archaeological Society and the Texas Memorial Museum, 1962. Thomas, D.H. “An Empirical Test for Steward's Model of Great Basin Settlement
Tiffany, Joseph A. and L. R. Abbott "Site Catchment Analysis: Applications to Iowa
Texas Press, 1987. Turner, Ellen Sue and Thomas R. Hester. A Field Guide to Stone Artifacts of Texas
Indians. 2nd edition. Lanham: Rowman and Littlefield, 1993.
Ethnology Bulletin No. 120, 1938. _____.Theory of Culture Change; the Methodology of Multi-linear Evolution. Chicago:
_____. The Archaeology and Bioarchaeology of the Gulf Coastal Plain: Volume II. Arkansas Archaeological Survey Research Series No. 38.,1990b.
Patterns.” American Antiquity 38 (1973): 155-176. _____.“Diversity in hunter-gatherer cultural geography.” In: Quantifying Diversity in
Archaeology, edited by Robert D. Leonard and George T. Jones. New York: Cambridge University Press, 1989.
Thünen, Johann Heinrich von. Isolated state; an English edition of Der isolierte Staat.
Translated by Carla M. Wartenberg. edited with an introd. by Peter Hall. Oxford, New York: Pergamon Press, 1966.
Archaeology", Journal of Field Archaeology 9 (1982): 313-322. Tull, Delena. Edible and Useful Plants of Texas and the Southwest. Austin: University of
Vayda, A.P. and B.J. McCay. “New Directions in Ecology and Ecological Anthropology.”
Annual Review of Anthropology 4 (1975): 293-305. Vayda, A.P. and R.A. Rappaport. “Ecology, Cultural and Non-Cultural.” In: Introduction
to Cultural Anthropology, edited by J.A. Clifton. Boston: Houghton, 1968. Verhagen, P. “The use of GIS as a Tool for Modelling Ecological Change and Human
Occupation in the Middle Aguas Valley (S.E. Spain).” In: Interfacing the Past, edited by Kamermans, H. and Fennema, K. Leiden: University of Leiden, Analecta Praehistorica Leidensia, 1996.
Vita-Finzi, C. Archaeological Sites in Their Setting, London: Thames and Hudson, 1978.
300
Vita-Finzi, C. and E.S. Higgs. “Prehistoric Economy in the Mount Carmel Area of Palestine: Site catchment analysis.” In: Proceedings of the Prehistoric Society 36 (1970):1-37.
Webley, D. Soils and Site Location in Prehistoric Palestine. In: Papers in Economic
Prehistory : Studies by members and associates of the British Academy Major Research Project in the Early History of Agriculture, edited by E.S. Higgs. Cambridge: Cambridge University Press, 1972.
Wheatley and Gillings. “Sites,Territories and Distances.” In: Spatial technology and
Archaeology : the archaeological applications of GIS. New York: Taylor & Francis, 2002.
Willmsen, E.N. “Interaction, Spacing Behavior, and the Organization of Hunting Bands.”
Journal of Anthropological Research 29 (1973):11-31. Zubrow, E.B.W. Prehistoric Carrying Capacity: A Model. Menlo Park: Stanford University.
Cummings Publishing Company, 1975.
_____.“Knowledge Representation and Archaeology: A Cognitive Example Using GIS.” In: The Ancient Mind: Elements of Cognitive Archaeology, edited by Renfrew, C. & E.B.W. Zubrow. Cambridge: Cambridge University Press, 1994.
301